Upload
others
View
6
Download
0
Embed Size (px)
Citation preview
Clemson UniversityTigerPrints
All Dissertations Dissertations
1-2010
PRODUCTION AND CHARACTERIZATIONOF NOVEL AIR FILTRATION MEDIAElizabeth SkomraClemson University, [email protected]
Follow this and additional works at: https://tigerprints.clemson.edu/all_dissertations
Part of the Materials Science and Engineering Commons
This Dissertation is brought to you for free and open access by the Dissertations at TigerPrints. It has been accepted for inclusion in All Dissertations byan authorized administrator of TigerPrints. For more information, please contact [email protected].
Recommended CitationSkomra, Elizabeth, "PRODUCTION AND CHARACTERIZATION OF NOVEL AIR FILTRATION MEDIA" (2010). AllDissertations. 683.https://tigerprints.clemson.edu/all_dissertations/683
PRODUCTION AND CHARACTERIZATION OF NOVEL AIR FILTRATION MEDIA
________________________________________________________________
A Dissertation Presented to
the Graduate School of Clemson University
____________________________________________________________________
In Partial Fulfillment
of the Requirements for the Degree Doctor of Philosophy
Materials Science and Engineering
____________________________________________________________________
By
Elizabeth Kristina Skomra May 2011
____________________________________________________________________
Accepted by:
Dr. Philip Brown, Committee Chair Dr. Christine Cole
Dr. Christopher Cox Dr. Deborah Lickfield
Dr. Olin Mefford
ii
ABSTRACT
HEPA and ULPA filtration systems have proven to be an advantageous instrument in
removing common contaminants from the air. However, an increased pressure drop due to
the build-up of particulates on the filters results in its inevitable decrease in performance.
Improving current filtration systems would include increasing collection efficiency all the
while either maintaining or reducing the differential pressure drop in order to extend the life
of the filter. One method of improving collection efficiency would be viable by increasing
the amount of surface area within the filter media by glass fibers because of their inherent
quality of being smaller in diameter offering more surface area than melt spun fibers. This
research examined alternate methods of producing fibers comparable or smaller in size than
glass fibers. As well, a unique geometry fiber know as a Capillary Channel Polymer (CCP™)
was examined for its contribution towards filtration since it offers at least twice the surface
area as a round fiber of equal denier.
Nonwoven filter media were manufactured with CCP™ fibers and tested for collection
efficiency and pressure drop. Although SEM images showed salt particles collecting within
the grooves of the shaped fibers, they did not exhibit HEPA quality efficiencies. The
pressure drop of these filters was low as compared to currently used M98 HEPA filters. This
was potentially due to the CCP™ fibers being unable to pack as closely together as round
glass fibers allowing for high air permeability which may have contributed to the lower
collection efficiency and pressure drop. Modified melt blown round fibers were also
iii
examined since their fiber diameters measured within nano range and offered benefits in
terms of ease of manufacturing. The nonwovens demonstrated HEPA quality collection
efficiency but at a higher pressure drop than M98 media. The melt blown nonwovens, in
addition to being thicker than the M98 media, lacked structural integrity which would allow
them to be used alone as a filter. The effect of slip flow on fibers measuring less than 0.50
µm in diameter was analyzed for M98 and meltblown media. The meltblown sample which
contained a higher amount of fibers within the slip flow regime and contained no scrim
demonstrated HEPA quality collection efficiency when compared to the M98 media with
comparable basis weight.
Dissolvable bi-component fibers were also examined for their potential to produce nano-size
sea fibers separated by a wet-laid process. Bi-component fibers can be manufactured via
traditional melt-spun lines and offer not only nano-size islands in round but also unique
geometry cross-sections such as CCP™. Difficulties in effectively dissolving off the polymer
sea leaving behind individual islands prevented an in-depth examination of their contribution
towards filtration. Composite media composed of CCP™ and meltblown layers proved
unsuccessful in terms of collection efficiency as well as thickness but demonstrated low
pressure drop. Further investigation into layering techniques and adding additional
meltblowns may prove fruitful for filtration media.
iv
ACKNOWLEDGEMENTS
I would like to take the opportunity to acknowledge the individuals responsible for helping
me to complete this dissertation and my graduate studies at Clemson University. First, I
would like to thank John Larzelere, Director of COLPRO S&T Research, for the funding,
guidance and insight for this project. To Dr. Philip Brown, my advisor, thank you for your
extensive knowledge of fiber science and contribution to my research and educational
studies. Thank you to Dr. Kathryn Stevens for her assistance in proofreading and feedback
on my writing and in numerous adventures in nonwoven production. A special thank you to
Dr. Chris Cox and Patrick Buckingham for their contribution to the modeling simulation
studies involved in this project. Thank you to Kim Ivey and all that work within her lab for
assistance in analytical testing. To all the AMRL EM Microscopy people, Dr. JoAn Hudson,
Taghi Darroudi, Donald Mulwee, and especially Dayton Cash, a huge thank you for not only
all the time and effort you put into helping with microscopy but making it a pleasant
experience as well.
I would like to acknowledge Bob Bowen, Kathy Bolton, Robbie Nicholson, and Dr. Chris
Cole, who all played a positive role in my experience over the past four years. Also, thank
you to Dr. Deborah Lickfield and Dr. Gary Lickfield, for going above and beyond in taking
time to help with my studies, particularly during my formative time at Clemson.
To my colleagues within Dr. Brown‘s lab group, a special acknowledgement for your
tutoring in science, your contributions to my research, and daily lab drama which created
memories which I will never forget. In particular, I would like to say thank you to Jessica
v
Domino, Dr. Lisa Fuller, Katelyn Howay, Julien Boyon, Brett Ellerbrock, Stephen Hipp,
Cody Reynolds, and Joel Barden. You guys are the best! Thank you, also, to my Mom and
Dad, family, and friends back in Michigan for always being supportive.
Last but certainly not least, to my husband, Scott Butler, for your constant support, love,
patience, kind words, and willingness to open a good bottle on a bad day, I say thank you
with all my heart!
vi
TABLE OF CONTENTS ABSTRACT .......................................................................................................................................................... ii
ACKNOWLEDGEMENTS ............................................................................................................................ iv
LIST OF FIGURES .........................................................................................................................................viii
LIST OF TABLES ............................................................................................................................................xiv
LIST OF EQUATIONS ................................................................................................................................... xv
INTRODUCTION ............................................................................................................................................. 1
1.1 History of Filtration .................................................................................................................................. 1
1.2 Filtration Theory and Modeling .............................................................................................................. 5
1.3 Fiber Manufacturing for Filters ............................................................................................................... 8
1.4 Filter Manufacturing ...............................................................................................................................19
1.5 Impact of Fiber Cross Sectional Geometry ........................................................................................23
1.6 HEPA and ULPA Filter Design ...........................................................................................................26
1.7 Mechanisms of Filtration .......................................................................................................................31
1.8 Aim of Research ......................................................................................................................................34
EXPERIMENTAL ............................................................................................................................................37
2.1 Materials ....................................................................................................................................................37
2.2 Fiber Extrusion ........................................................................................................................................39
2.3 Bi-component Dissolution.....................................................................................................................40
2.4 Nonwoven Manufacturing .....................................................................................................................40
2.5 Characterization of Fibers ......................................................................................................................42
2.6 Filter Characterization ............................................................................................................................46
2.7 Modeling Simulation ...............................................................................................................................51
RESULTS and DISCUSSION .........................................................................................................................52
3.1 Benchmarking of M98 Filter Media .....................................................................................................52
3.1.3 M98 collection efficiency ....................................................................................................................65
3.2 Characterization of Capillary Channel Polymer™ Fibers .................................................................67
3.3 Modeling of CCP™ Fibers ....................................................................................................................79
3.4 Round Fiber versus CCP™ Fiber Filter Media ..................................................................................91
3.5 Meltbown Fibers ................................................................................................................................... 102
3.6 Bi-component Fibers ........................................................................................................................... 124
3.7 Composite Filter Media ....................................................................................................................... 157
vii
CONCLUSIONS ............................................................................................................................................ 165
4.1 Capillary Channel Polymers (CCP™) Fibers for Use in HEPA and ULPA Filtration ............. 165
4.2 Meltblown Nonwovens for Use in HEPA and ULPA Filtration ................................................. 166
4.3 Bi-component Fibers for Use in HEPA and ULPA Filtration ..................................................... 167
4.4 Composite Filter Media ....................................................................................................................... 168
FUTURE WORK ........................................................................................................................................... 170
5.1 Unique Geometry Fibers ................................................................................................................ 170
5.1.1 Unique geometry fibers through meltblown process.............................................................. 171
5.2 Electrostatic charge ......................................................................................................................... 171
5.3 Modeling Simulation ....................................................................................................................... 173
APPENDIX A ................................................................................................................................................. 174
REFERENCES ............................................................................................................................................... 198
viii
LIST OF FIGURES
Figure 1.1 Schematic of electrospinning system ............................................................................ 10
Figure 1.2 Electrospun acrylic nano-fibers at 1,100x magnification demonstrating fine
........diameter size................................................................................................................................ 10
Figure 1.3 Schematic of meltblown nonwoven line ...................................................................... 13
Figure 1.4 Meltblown fiber substrate sample produced by Hills Inc. at 2,000x magnification
........showing variability in fibe size.................................................................................................. 14
Figure 1.5 Schematic of bi-component melt line .......................................................................... 17
Figure 1.6 Various bi-component fibers manufactured by Hills Inc. ......................................... 18
Figure 1.7 Geometry of Y and triple Y fibers from Cornell University study on collection
........efficiency a) Y fiber b) Y fiber in reversed orientation c) tripleY and d) triple Y fiber in
........reversed orientation54 ................................................................................................................. 24
Figure 1.8 Cross-sectional image of CCP™ fiber ......................................................................... 25
Figure 1.9 Different pore types in filtration media ....................................................................... 29
Figure 1.10 Key mechanisms of filtration during HEPA and ULPA filtration ........................ 32
Figure 1.11 Schematic of collection efficiency vs. particle size of most penetrating particle
........size (MPPS)13 ............................................................................................................................... 33
Figure 2.1 Schematic of Buchner funnel used for wet-laid filter media production ................ 41
Figure 2.2 Hills microtone used during cross-sectional fiber preparation for optical
........microscopy .................................................................................................................................. 42
Figure 2.3 Embedding cap with stand used for cross-sectional fiber preparation ................... 43
Figure 2.4 Example of SEM image with scope, beam current (kV), working distance,
........magnification, detector type, vacuum setting (Pa), and time information ......................... 45
Figure 2.5 Schematic of Capillary Flow Porometer ...................................................................... 47
Figure 3.1 SEM image taken at the end of the edge of a cross-section of M98 filter media .. 54
Figure 3.2 SEM image of cross-section of M98 filter media where fibers appear to be in
........sheet-like layers ........................................................................................................................... 54
Figure 3.3 SEM image of exterior layer of M98 filter media at 110x magnification ................ 55
Figure 3.4 SEM image of exterior layer of M98 filter media at 8,000x magnification showing
........variability of fiber diameter as well as polymer binder film ................................................. 55
Figure 3.5 SEM imageof inner layer of M98 filter media at 60x magnification where larger
........fibers appear to be laid in a light netting effect ..................................................................... 56
Figure 3.6 SEM image of interior layer of M98 filter media at 9,000x magnification showing
........fibers varying in diameter .......................................................................................................... 56
Figure 3.7 Fiber size combined distributions of M98 polymer and glass fibers ....................... 58
ix
Figure 3.8 ATR-IR spectra of M98 filter media post and pre TGA showing characteristic
........glass peaks only post TGA and peaks of PET and glass prior to TGA ............................ 59
Figure 3.9 TGA thermogram of M98 filter media ........................................................................ 60
Figure 3.10 M98 filter media composed of only glass fibers after pryolysis in TGA removal
........of polymer fibers at 400x magnification ................................................................................. 62
Figure 3.11 M98 filter media composed of only glass fibers after pryolysis in TGA removal
........of polymer fibers at 1000x magnification ............................................................................... 62
Figure 3.12 M98 filter media composed of only glass fibers after pyrolysis in TGA removal
........of polymer fibers at 9,000x magnification .............................................................................. 63
Figure 3.13 M98 glass fiber size distributions after pyrolysis in TGA to remove polymer
........fibers ............................................................................................................................................. 63
Figure 3.14 Fiber size distribution of M98 filter media pre-TGA and post-TGA ................... 64
Figure 3.15 Comparison of M98 as received and post-TGA fiber size diameters ................... 64
Figure 3.16 Collection efficiency vs. flow rate of M98 filter media at 0.3 µm particle size .... 65
Figure 3.17 Collection efficiency vs. pressure drop of M98 filter media at 0.3 µm particle size
Llll.ll.. .................................................................................................................................................... 66
Figure 3.18 Collection efficiency vs. flow rate of M98 filter media at MPPS (µm) ................. 67
Figure 3.19 Relative size of common materials and common air contaminants13 .................... 68
Figure 3.20 Optical image of 30 dpf polypropylene CCP™ fiber demonstrating shape
........retention....................................................................................................................................... 70
Figure 3.21 Scanning electron microscope image of bi-component fiber composed of
........EVOH sea and Nylon 6 island demonstrating shape retention .......................................... 70
Figure 3.22 Scanning electron image of bi-component fiber composed of PLA sea and
........polypropylene islands which is not demonstrating correct shape integrity ....................... 71
Figure 3.23 3 dpf CCP™ fiber with groove measuring 8-9 µm.................................................. 72
Figure 3.24 6 dpf CCP™ fiber with grooves measuring 12 µm and 16. - 17 µm .................... 73
Figure 3.25 15 dpf CCP™ fibers with grooves measuring 17 µm. ............................................. 73
Figure 3.26 20 dpf CCP™ fiber grooves measuring 17 µm and 12 µm .................................... 74
Figure 3.27 30 dpf CCP™ fiber with groove measuring 24 µm ................................................. 74
Figure 3.28 Bi-component CCP™ fiber pre-dissolution of the sea, grooves measuring
........around 2 µm .............................................................................................................................. 75
Figure 3.29 Bi-component CCP™ fibers post-dissolution of the sea ........................................ 75
Figure 3.30 CCP™ fiber loaded with salt particles at 2,000x magnification ............................. 77
Figure 3.31 CCP™ fiber loaded with salt particles at 5,000x magnification ............................. 77
Figure 3.32 Single fiber efficiency of round fibers ........................................................................ 80
Figure 3.33 Single fiber efficiency for CCP™ fiber ...................................................................... 81
Figure 3.34 Single fiber efficiency results of round versus CCP™ fiber with varying particle
........density at a fixed radius of 0.2 µm ........................................................................................... 82
x
Figure 3.35 Single fiber efficiencies of round and CCP™ fibers in horizontal and vertical
........orientations.................................................................................................................................. 83
Figure 3.36 Single fiber efficiency results of round versus CCP™ fiber with varying particle
........density and fixed Stoke‘s number of 0.3 ..........................................................................................84
Figure 3.37 SEM cross-section image of nonwoven filter media composed of CCP™ fibers,
........of varying denier, and low melting PET binding fibers used for simulation study ...........85
Figure 3.38 Constructed simulation domain based off of SEM images of nonwoven filter
........media composed of CCP™, of varying denier, and low melting PET binding fibers ......86
Figure 3.39 Comparison of simulation to experimental data of collection efficiency of
........CCP™ filter media .................................................................................................................................87
Figure 3.40 Comparison of simulation to experimental average of collection efficiency of
........CCP™filter media ..................................................................................................................................87
Figure 3.41 Simulated pressure drop of CCP™ fiber filter media in Pascals ...............................89
Figure 3.42 Simulated pressure drop of round fiber filter media in Pascals .................................89
Figure 3.43 Simulation of flow path of particles in a filtration media composed of CCP™
........fibers ............................................................................................................................................. 90
Figure 3.44 Simulation of flow path of particles in a filtration media composed of round
........fibers ............................................................................................................................................. 91
Figure 3.45 Comparison of flow rate (L/Min) versus differential pressure (PSI) of 8 layered
........15 dpf CCP™ and round nonwoven filter media ........................................................................93
Figure 3.46 Comparison of pressure drop of CCP™ versus round fiber filter media with
........normalized weights .................................................................................................................................94
Figure 3.47 Comparison of flow rate (L/Min) versus differential pressure (PSI) of filter
........media made from 3 layers of 15 dpf CCP™ and round fibers with comparable surface
........area ..............................................................................................................................................................96
Figure 3.48 Comparison of flow rate (L/Min) versus differential pressure (PSI) of filter
........media made from 4 layers 15 dpf CCP™ and round fibers with comparable surface area
llllllll ......................................................................................................................................................................96
Figure 3.49 Collection efficiency of filter media composed of 80% 15 dpf CCP™ fibers .....97
Figure 3.50 Pressure drop of filter media made from 80% 15 dpf CCP™ fibers.......................98
Figure 3.51 Collection efficiency of filter media composed of varying sized CCP™ fibers 100
Figure 3.52 Collection efficiency of CCP™ and M98 media at varying particle diameters an
........flow rates ................................................................................................................................................ 101
Figure 3.53 Fiber size distribution graphs for meltblown fibers manufactured by Hills Inc
lllllll .................................................................................................................................................................... 104
Figure 3.54 111808-01 at 2,000x magnification showing overall construction on nonwoven
........material .................................................................................................................................................... 105
Figure 3.55 111808-01 at 2,200x magnification showing detail of fiber size variability ......... 105
Figure 3.56 111808-01 at 11,000x magnification detailing fiber size differences ..................... 106
xi
Figure 3.57 111808-01 fibers measuring 63 and 184 nm ................................................................. 106
Figure 3.58 111808-02 at 1,500x magnification showing variability in fiber diameters .......... 107
Figure 3.59 111808-02 fiber measuring 180 nm ................................................................................. 108
Figure 3.60 111808-03 with large backing fibers (~17-22 µm) ...................................................... 109
Figure 3.61 111808-03 at 5,000x magnification ........................................................................... 109
Figure 3.62 111808-03 at 70,000x magnification with fiber measuring 80 nm ......................... 110
Figure 3.63 111808-04 at 450x magnification with backing fibers visible in foreground ...... 111
Figure 3.64 111808-04 at 500x magnification with backing fibers in background ................... 111
Figure 3.65 111808-04 at 4,000x magnification .................................................................................. 112
Figure 3.66 111808-06 at 2,500x magnification with varying fiber sizes ..................................... 113
Figure 3.67 111808-06 with fiber measuring at 161 nm ................................................................... 113
Figure 3.68 Collection efficiency versus flow rate for 111808-xx single layer meltblown
........samples .................................................................................................................................................... 115
Figure 3.69 Percentage of meltblown fibers measured less than 0.5 µm in size ....................... 117
Figure 3.70 Collection efficiency versus pressure drop for 111808-xx single layer meltblown
........samples .................................................................................................................................................... 118
Figure 3.71 Collection efficiency of 111808-06 multi layered meltblown samples ................ 119
Figure 3.72 Collection efficiency versus pressure drop for 111808-06 layered samples at 8.3
........L/Min flow rate ........................................................................................................................ 120
Figure 3.73 Collection efficiency of 111808-06 meltblown sample versus M98 filter media at
........8.3 L/Min flow rate ................................................................................................................. 121
Figure 3.74 Thickness of single layer meltblown sample 111808-06 measuring 362 µm ...... 123
Figure 3.75 Thickness of M98 filter media measuring 600 µm ................................................. 123
Figure 3.76 Comparison of fiber size distribution of meltblown to M98 filter media .......... 124
Figure 3.77 IR spectra of bi-component 40/60 PET/WSPET snowflake fiber .................... 127
Figure 3.78 IR spectra of bi-component 156,000 islands-in-a-sea fiber composed of 80/20
ESPET sea/Nylon 6 island............................................................................................................. 128
Figure 3.79 TGA analysis of 40/60 WSPET/PET snowflake fibers heated at 20°C per
........minute ........................................................................................................................................ 129
Figure 3.80 TGA analysis of 10/90 Nylon 6/ESPET 156,000 islands-in-a-sea fiber heated at
........20°C per minute ....................................................................................................................... 130
Figure 3.81 DSC thermogram of 40/60 PET/WSPET snowflake fiber before quench ...... 131
Figure 3.82 DSC thermogram of 40/60 PET/WSPET snowflake fiber after quench.......... 132
Figure 3.83 DSC thermogram of 10/90 Nylon 6/ESPET 156,000 islands-in-the-sea fiber
........before quench ........................................................................................................................... 133
Figure 3.84 DSC thermogram of 10/90 Nylon 6/ESPET 156,000 islands-in-the-sea fiber
........after quench ........................................................................................................................................... 134
Figure 3.85 CCP™ PP island/PLA sea bi-component fiber without shape integrity ............ 135
xii
Figure 3.86 Nylon 6 CCP™ island/Poly(vinyl alcohol) sea fiber cross section at 3,200x
........magnfication pre-dissolution of the sea ........................................................................................ 136
Figure 3.87 Nylon 6 CCP™ fibers post dissolution at 2,700X magnification .......................... 137
Figure 3.88 Snowflake WSPET sea fibers prior to any attempt to dissolve out the sea ........ 138
Figure 3.89 Snowflake WSPET sea fibers after 2 hours in 60°-70°C water treatment .......... 139
Figure 3.90 Snowflake WSPET sea fibers after being immersed in 80°-70°C water for
........approximately 3.5 hours..................................................................................................................... 139
Figure 3.91 Snowflake WSPET sea fibers at 300x magnification after exposure to solvents
........and sonicator for 2.5 hours ............................................................................................................... 141
Figure 3.92 Snowflake WSPET sea fibers at 470x magnification after exposure to solvents
........and sonicator for 2.5 hours ............................................................................................................... 142
Figure 3.93 Snowflake PET island fiber measuring 210nm after exposure to solvents and
........sonicator for 2.5 hours ....................................................................................................................... 142
Figure 3.94 Snowflake 20/80 Nylon 6/ESPET fibers sea after 3 cycles in 27% NaOH ...... 144
Figure 3.95 Snowflake 20/80 Nylon 6/ESPET sea after 3 cycles in 27% NaOH and 8%
........butyl alcohol .......................................................................................................................................... 145
Figure 3.96 Snowflake Nylon 6 islands measuring 228.6 nm ......................................................... 145
Figure 3.97 156,000 islands-in-the-sea bi-component fiber at 1,500x magnification where sea
........is not dissolved off but islands are exposed ................................................................................ 147
Figure 3.98 156,000 islands-in-the-sea bi-component fibers at 1,800x magnification where
........sea is not dissolved off but islands are emerging ....................................................................... 147
Figure 3.99 156,000 islands-in-the-sea bi-component fibers at 4,000x magnification where
........sea is not dissolved off but islands are visible ............................................................................. 148
Figure 3.100 Single ‗island‘ in 156,000 islands-in-the-sea bi-component fibers measuring 44 --
........nm ............................................................................................................................................................. 148
Figure 3.101 156K islands-in-the-sea fibers after 1 day in sonicator and solvents .................. 151
Figure 3.102 156K islands-in-the-sea after 5 days in sonicator (non-continuous) and solvents
llllllll .................................................................................................................................................................... 151
Figure 3.103 156K islands-in-the-sea after 7 days in sonicator (non-continuous) and solvents
llllllll .................................................................................................................................................................... 152
Figure 3.104 156K islands-in-the-sea after 3 days in sonicator (continuous) and solvents
llllllll .................................................................................................................................................................... 152
Figure 3.105 156K islands-in-the-sea after 5 days in sonicator (continuous) and solvents
llllllll .................................................................................................................................................................... 153
Figure 3.106 156K islands-in-the-sea after 7 days in sonicator (continuous) and solvents at
........1,000x magnfication ............................................................................................................................ 153
Figure 3.107 156K islands-in-the-sea after 7 days in sonicator (continuous) and solvents at
........11,000x magnification......................................................................................................................... 154
xiii
Figure 3.108 156K islands-in-the-sea after after 32 hours in solvent with the addition of
........beznyl alcohol ....................................................................................................................................... 155
Figure 3.109 156K islands-in-the-sea after 56 hours in solvent with the addition of benzyl
........alcohol ..................................................................................................................................................... 156
Figure 3.110 156K islands-in-the-sea after 68 hours in solvent with the addition of benzyl
........alcohol ..................................................................................................................................................... 156
Figure 3.111 156K islands-in-the-sea after 92 hours in solvent with the addition of benzyl
........alcohol ..................................................................................................................................................... 157
Figure 3.112 Schematic of composite filter media composed of CCP™ and meltblown layers
llllllll .................................................................................................................................................................... 158
Figure 3.113 Collection efficiency of composite filter media with varying flow rates
llllllll .................................................................................................................................................................... 161
Figure 3.114 Pressure drop of composite filter media ...................................................................... 161
Figure 3.115 Figure of merit measurements of meltblown samples versus M98 filter media
llllllll .................................................................................................................................................................... 164
Figure 5.1 PP core/PLA sheath bi-component fiber for potential investigation for use in
........filtration media ...................................................................................................................................... 171
Figure 5.2 Schematic of electrostatically charged nonwoven filter media .................................. 172
xiv
LIST OF TABLES
Table 2.1 List of materials used during experiments ............................................................................37
Table 2.2 List of melt spun, bi-component, and meltblown fibers investigated for use as
........HEPA and ULPA filtration media as well as M98 filter media used as a benchmark
........material .......................................................................................................................................................38
Table 3.1 Group frequency assignments for glass, PET,and polystyrene ....................................60
Table 3.2 Shape factor for round and CCP™ fibers ............................................................................78
Table 3.3 Varying density of particles in kilograms per cubic meter of round versus CCP™
........single fiber efficiency modeling ..........................................................................................................82
Table 3.4 Input parameters for modeling pressure drop of filter media made of CCP™ fibers
........to round fibers ............................................................................................................................ 88
Table 3.5 Surface area and basis weights of 15 dpf round fiber filter media and 15 dpf
........CCP™ fiber filter media ........................................................................................................... 95
Table 3.6 Filter composition of CCP™ filter media varying in denier ...................................... 99
Table 3.7 Meltblown nonwovens manufactured by Hills Inc. .................................................. 102
Table 3.8 Collection efficiency and resistance results of meltblown sample versus M98 filter
........media .......................................................................................................................................... 121
Table 3.9 List of bi-component fibers analyzed for use in filter media ................................... 125
Table 3.10 Dissolution solvents used to attempt to dissolve off WSPET sea and PET islands
llllllll .................................................................................................................................................... 149
Table 3.11 Dissolution solvents used to attempt to dissolve off ESPET sea with Nylon 6
........islands ......................................................................................................................................... 143
Table 3.12 Dissolution parameters used to attempt to dissolve off sea of 156K islands-in-
........the-sea fibers ............................................................................................................................. 148
Table 3.13 Dissolution parameters used to attempt to dissolve off sea of 156K islands-in-
........the-sea fibers using benzyl alcohol ........................................................................................ 154
Table 3.14 Composition of composite filter media made with CCP™ fibers and meltblown
........nonwovens ................................................................................................................................ 159
Table 3.15 Descriptionof polymer, shape, and basis weights of FOM filter media ............... 163
Table 3.16 Composition of FOM filter media ............................................................................. 163
xv
LIST OF EQUATIONS
Equation 1 Stoke‘s Law ....................................................................................................................... 5
Equation 2 Knudsen‘s Flow ............................................................................................................... 7
Equation 3 Shape Factor ................................................................................................................... 78
Equation 4 Stoke‘s Equation ............................................................................................................ 79
Equation 5 Single Fiber Efficiency for Round Fiber .................................................................... 80
Equation 6 Single Fiber Efficiency for Shape Fiber ..................................................................... 80
Equation 7 Cross-sectional area conversion for 3D to 2D modeling ........................................ 84
Equation 8 Darcy‘s Law .................................................................................................................... 98
Equation 9 Figure of Merit (FOM) ............................................................................................... 162
1
INTRODUCTION
As ambient air becomes more polluted with bacteria, molds, dust, smoke, pollen, and
carbon, the need to filter out air born pollutants becomes more vital to the health of the
general population as asthma, allergies, and respiratory symptoms increase. These concerns
have led to a demand for products that can improve air quality 1. High efficiency particulate
air (HEPA) and ultra low penetration air (ULPA) filters are used in a wide variety of
industries, from clean room applications to vacuum cleaner bags. A HEPA filter is defined
as a filter with an efficiency of 99.97% or greater for 0.3 µm diameter particles. ULPA filters
are defined as having an efficiency of 99.999% or greater against 0.1-0.2 µm particles2.
Generally, these filters are composed of nonwoven materials made of glass fibers with round
cross sectional geometry. Perhaps the two most important features of air filters are pressure
drop and particle collection efficiency. Pressure drop can be defined as the difference in
static pressure between the upstream and downstream side of the filter and increases as the
filter becomes loaded with particles3. Both pressure drop and collection efficiency depend on
operating conditions, filter structure, and the types of particles being collected4. An ideal
filter would have high particle collection efficiency with minimum pressure drop during
operation.
1.1 History of Filtration
Earliest forms of air filters date back to Roman times at which time they utilized ‗loose
bladders‘ to protect workers from mine dust and ‗industrial dust‘5. In the mid 1450s,
Leonardo da Vinci spoke about the use of wet clothes to prevent against inhalation of
2
warfare fumes, presumably sulfur. In 1560, Renaissance physician Paracelsus was the first to
write about the associated health concerns of dust when he spoke about ‗dust diseases‘ in his
book ‗Von der Bergkrankheiten und anderen’. The first account of a filtration device for
respiration was by Brisé Faden in 1814. Fraden‘s device was simply a cotton filled box which
attached to the mouth via a hose6.
Literature evidence supports filtration advancement in the 1800s due to issues with smoke,
chemicals, and medical needs. The Industrial Revolution and the necessity to burn coal for
fuel became an efficient way to facilitate growth of cities but also increase the amount of
smog and smoke7. In 1823, scientists John and Charles Dean developed a firefighter‘s air
purifying mask that reduced the amount of acrid smoke and chemical fumes for the wearer.
This was a significant improvement on the prevalent wet cloth around the mouth method
which could not provide adequate protection for the firefighters at that time. The essence of
this smoke helmet was that is supplied fresh air via a hose using an air pump that was
outside the zone of toxicity. In 1854, John Stenhouse took this basic design and updated it
for divers and coal miners by including charcoal, a design which is very similar to the carbon
filters still being used today. By 1871, John Tyndall upgraded the original firefighters mask
by introducing a respirator, making it the first portable filtration system.8, 9. In addition,
Professor Tyndall, a physicist, was the first to qualitatively test the performance of aerosol
filters based on a ‗light-scattering‘ observations of smoke particles. Motes in the air were
illuminated by an electric lantern and the paths of the rays were marked and studied10.
The earliest forms of high efficiency air filtration systems were introduced by the Germans
with military gas masks that filtered out submicron particles. It was not until World War II
3
that a classified plan known as the Manhattan Project was responsible for the development
and term HEPA (High Efficiency Particulate Air) filter which was coined by Humphrey
Gilbert in his 1961 Atomic Energy Commission Report11. Due to the Manhattan Project and
atomic bomb, the US Atomic Energy Commission needed to develop a filter to protect
soldiers and scientists from radioactive particles on the battlefield. Although proven
ineffective against atomic radiation, HEPA filter offered limited protection against chlorine
gas, mustard gas, and smoke12. After World War II, HEPA filters found increased military
use since they were required to protect operational headquarters where gas masks were not
practical. Shortly thereafter, the project was declassified and the HEPA filter received
various make-overs, including the introduction of ULPA filters in 1961. In the 1970s and
80s HEPA filter usage increased due to civilian consumer concerns with air pollution.
Initially, HEPA purifying systems were large and bulky and used mainly in hospitals,
pharmaceutical industries, and computer chip manufacturers. More recently they have
become more popular in residential and industry filters to protect against allergens. Through
the years, various air filtration companies have put their own technological spin on HEPA
and ULPA filters making them into the commercial product we see today 13,9,8.
1.1.1 History of Fibers in HEPA Filtration
Asbestos fibers mixed with esparto pulp were the first raw materials to be used in HEPA
filtration. Asbestos is a general term for six naturally occurring minerals which are all now
considered dangerous. They are fiber bundles made of extremely long and thin fibers which
are easily separated. Use of asbestos dates back to Roman times where it was found in
cremation cloths to separate the ashes of upper class from the slaves as well as in lamp
4
wicks. In addition, it has been found in the armor of knights from the Middle Ages where it
acted as insulation. Asbestos can be divided into two groups, serpentine and amphibole,
which have different physical characteristics. Chrysotile, white asbestos, and crocidolite, blue
asbestos, were the prevalent two found in HEPA filters. Chrysotile is the only serpentine
form and is unique because of its tiered curly, flexible, fibers which make it possible to be
spun and woven into fabrics and is found in 90% of asbestos products. Crocidolite is an
amphibole form of asbestos and is made of needle-like fibers. Beginning in the 1800s,
asbestos was mined commercially in Canada and soon found use in United States in steam
locomotives by reducing heat build-up and temperature fluctuations. Soon thereafter, it was
incorporated into train brakes and clutches and other products which needed high tensile
strength, chemical, thermal, and electrical resistance13, 14.
Asbestos was also popular choice for military gas masks used during World War II, due to its
fine nano-sized fiber diameters, ease of processing, in addition to being heat and chemical
resistant. Since these raw materials were imported, the US government funded research and
development of new media due to fear of shortage. This later proved beneficial because of
the recognition of known carcinogenic health hazards. Although there was speculation for
years over health hazards, it was not until the 1970s that there was solid evidence of health
issues such as asbestosis, lung cancer, and mesothelioma from these submicron fibers and
soon asbestos was mandated out of over 3,000 products13, 14.
In the mid 1950s, Arthur D. Little introduced glass microfibers filter media. Glass fibers are
still used in HEPA filters today as they have similar physical properties to asbestos fibers,
i.e., they are of fine diameter and have high heat and chemical resistance Fine fiber diameters
5
provide increased surface area and produces a filter with better collection efficiency which
made both these fiber types ideal for filtration2.
Advancements in melt spinning and meltblown technology has helped broaden the list of
raw materials for use in nonwoven filters since fibers in the sub-micron and even nano scale
range can be produced. Nylon, polyethylene, polyester, and polypropylene are just a few of
the synthetic organic fibers used today13. Also, synthetic polymers give greater flexibility for
tailoring filtration characteristics as compared to glass fibers. Polymers can be
electrostatically charged which helps aid in filtration. For example, Turnhout, J.V. et al.15
proved polymer fibers turned into electrets filter have higher cleaning efficiency with low
resistance and a relative large dust-holding capacity. The electrets fibers are charged via a
coronoa plasma treatment which makes them dielectrics carrying a strong positive and
negative electric charge. The advantage this imparts over mechanical filtration is that the
fibers are effective in collecting particles when relatively far apart. This in turn allows filters
to be more open and this assists in reducing differential pressure. In addition, it has been
suggested that polymers may provide mechanical benefits such as high tensile strength and
good strain resistance as well as low overall environmental impact (in terms of energy use)16.
1.2 Filtration Theory and Modeling
Although there is no single filtration theory that explains the complexities of the fluid
dynamics of HEPA and ULPA filtration, most gas flow modeling is based on the Stoke‘s law
[equation 1].
c [equation 1]
6
where
F = Force R = radius of the sphere
y
c = velocity through a continuous fluid
The equation represents the force exerted on the sphere by the fluid and discusses the
motion of spheres through viscous fluids. Some assumptions must be made including13, 17:
Particle build up not effecting flow
Incompressible flow
Reynolds numbers << 1
Fibers are parallel
Flow across fibers
Cox18 uses the Stoke‘s equation as his governing equation in modeling a simulation of the
flow and capture of particles (filter efficiencies) in a fibrous filter media made from both
round and CCP fibers. These simulations were achieved through the use of grid generation,
finite element flow solver, and Brownian dynamic simulation19. Cox‘s modeling assumed
that air viscosity, air temperature, and particle density were fixed parameter values. The
model and experimental results will be discussed in detail in the results and discussion
chapter of this dissertation.
1.2.1 Knudsen Flow and Slip Flow Effect
Recent research has looked at the use of nano-fibers in filtration, not only for their ability to
increase collection efficiency but also for the potential of reducing pressure drop. This
theory is based on Knudsen‘s Flow [equation 2].
7
[equation 2]
where
K = Knudsen number λ = mean free path of a molecule ι = linear dimension of the flow field
Traditionally, Knudsen flow describes the movement of fluids with a high Knudsen number,
specifically, where the dimension of the flow space is the same or smaller order of
magnitude as the mean free path13, 20.
It has been suggested that the Knudsen number can be related to fiber diameter13, 21.
where
K = Knudsen number λ = mean free path of a molecule
= radius of the fiber
It is thought when Kn > 0.1, slip flow will prevail. For standard air conditions, the mean free
path is 0.066 microns, therefore, for fiber with diameters smaller than 0.5 microns, slip flow
must be considered22-24. This slip flow will predominate at the fiber surface which allows for
more air traveling near the fiber surface which leads to more particles traveling near the
surface increasing probability of particle capture. In addition, the increase of pressure drop
with decreasing fiber diameter is less steep for nano-fibers due to the slip effect.
8
Donaldson Company Inc., a worldwide provider of filtration systems, made a media
composed of electrospun fibers of ~ 250 nm in diameter on a cellulose substrate composed
of fibers ~ 10 µm in diameter. Their results stated they saw an increase in particle collection
efficiency without a corresponding decrease in filter life or significant increase in pressure
drop, fielding testing results also confirmed these results25. Donaldson attributed these
findings to slip flow effect.
1.3 Fiber Manufacturing for Filters
Since fiber diameter plays a large role in the effects of HEPA filter quality, the means by
which fiber size is achieved will be examined. Air filter media is commonly made from glass
or synthetic fibers such as polyester, polypropylene, polyamides, or acrylic binders. Glass
fibers have the advantage of being extremely fine (about 1.0 µm in diameter) which aid in
their packing efficiency, which leads to higher density filters. The disadvantage is that this
superior packing also increases pressure drop. On the other hand, synthetic fibers made
from traditional melt spinning techniques are usually larger in diameter which tends to make
more open filter media resulting in lower pressure drop. In addition, the larger fibers tend to
leave void spaces between them which can lead to higher capacity for particle capture but
also reduces the chance of a particle colliding with a fiber and adhering to it26. There are
several methods which produce sub-micron polymer fibers which are used for filter media.
Their advantages and disadvantages will be examined in the following sections.
1.3.1 Electrospinning
Lord Rayleigh theorized and later experimentally showed that an electrical charge of certain
size can cause a droplet at a nozzle tip to overcome surface tension and eject in a stream27.
9
Electrospinning is a method that makes use of Rayleigh‘s principle. In 1934, the first
electrospinning patent was granted for fine fibers made from cellulose acetate28. In
electrospinning, nano-scale fibers are produced in the form of a web from a polymer
solution in a needle. Long continuous filaments are ejected from the needle by overcoming
the surface tension of the solution droplet at the tip of that needle through electrostatic
charge. Figure 1 is a schematic of a basic electrospinning system. First, a voltage is initially
applied to a polymer solution and the droplet forms a hemispherical surface. With increasing
electrical field, the surface shape changes from hemispherical to spherical and finally to
conical which is called a Taylor cone. When the electrical charge is high enough that it
overcomes the surface tension, a stream is ejected from the Taylor cone onto a collecting
plate to make a fiber.
Electrospun fiber diameter is determined by numerous variables including polymer
molecular weight and chemistry, as well as solvent ratio in solution which in turn affects
viscosity, surface tension, and conductivity27. Such fibers have a very high surface area due to
the small diameter fibers and can produce media with very small pore sizes. These nano-
fibers have led to an improvement in the filter-like media performance, such as particle
capture29. Figure 2 shows a micrograph of an electrospun nano-fiber web with fiber
diameters on the order of 80 nm, which highlight the potential of electrospinning method to
produce filter like media.
10
Figure 1.1 Schematic of electrospinning system
Figure 1.2 Electrospun acrylic nano-fibers at 1,100x magnification demonstrating fine diameter size
11
Ahn, et al.29 have successfully developed nano filters made from Nylon 6 nano fibers using
this method. In addition, Kim et al.30 produced a nanofibrous membrane by electrospinning
polycarbonate (PC)/chloroform solution. These filters showed good filtration efficiency
when compared with a HEPA filter and exhibit a comparable pressure drop although do not
provide the depth of filtration that HEPA filters do. Reneker and Chun have demonstrated
that electrospinning fibers can also occur from melt in vacuum and air, with melt in vacuum
being advantageous because higher fields and temperatures can be used31. However,
electrospinning speeds, specifically solution spun rates up to 0. 1g/min onto a 1 meter wide
web, are slow compared to common industrial spinning processes10,32,110. In addition,
electrospinning in polymer solution frequently uses toxic solvents that have to be recovered
and disposed of properly33. There has been a push towards electrospinning from the melt
due to potentially faster production rates although fiber diameter tends to be larger when
compared to solution electrospinning32.
1.3.1.1Charge injection method
Electrohydrodynamic atomization of liquids by the charge injection method is an alternate to
the electrospinning method discussed and offer advantages in terms of output and efficiency.
An example of the set up, produced first by A.J. Kelly in 2000, consists of two electrodes
immersed in a non-conducting fluid with the sharpened point of the emitting electrode held
at a high electric potential centered over a grounded orifice. A small distance between the
electrodes exists, averaging around one to two orifice diameters, and an intense electric field,
much greater than in electrospinning, is set up in the fluid. Under high pressure, the fluid is
continuously forced through the orifice and it becomes highly charged as it passes between
12
the electrodes. The majority of the charge remains in the liquid due to the low mobility of
electrons in the insulating fluid and because of the short residence time of the fluid prior to
exiting the orifice. The flow rate is typically between 0.2 and 5 ml/s, around three times
greater than used in electrospinning. Once the droplets emerge from the orifice, they can be
collected on a third collector electrode in the form of a grounded or oppositely charged
object, or they can be allowed to disperse freely in the environment34.
Unlike electrospinning, no Taylor cone is formed and the velocity of the fluid stream is
determined by the mechanical pressure applied not by the strength of the external electric
field. This allows the charge injection electrospray method to overcome limitations by
decoupling the fluid flow from the field strength. While electrospinning is applicable to
conductive liquids, charge injection technology can only be used with insulating or weakly
conducting fluids. One disadvantage in using this method for ultra-efficiency filtration
properties is that the webs produced have a lower conversion rate into nanofibers. The
membranes have nanofibers with unconverted polymer or they are fully fibrillar but contain
a mixture of both nanofibers and microfibers34.
1.3.2 Meltblown
Metlbown fibers are extruded and drawn by high velocity heated air to make fine fibers in
the 1-4 µm diameter range. Figure 1.3 is a schematic of a meltblown line where the
thermoplastic filaments are cooled and collected onto a screen. Self-consolidation of the web
is common due a high level of filament entanglement. Since the fibers are tacky when
collected they bond together to form a web, at other times, bonding methods such as a
calendar rolls are used to further consolidate the media35. Meltblown nonwovens tend to
13
form softer and weaker webs as compared to other nonwoven processes such as spunbond.
Therefore, they are commonly used with a substrate for added structural intergrity13. Recent
advancements have allowed a modified meltblown process to produce fibers that measure
less than 1 µm in diameter. The meltblown process has the advantage of producing fibers at
a much faster rate than electrospinning. Hills Inc. has demonstrated mass rates of 10g per
minute on a 50cm wide belt, Appendix A. Due to this fast production rate of submicron
fibers it makes this meltblown process a valuable method of filter media production. A
negative attribute of meltblowns is their inherent weak nature. Based on the requirements of
the end product, it may be necessary to layer the final product (upon itself) or on a substrate
scrim resulting in increased production cost36.
Figure 1.3 Schematic of meltblown nonwoven line
14
Figure 1.4 is an image of a meltblown nonwoven produced by Hills Inc. An attribute of
meltlbown materials that must be considered is the variability in fiber diameter which may
need to be taken into account when designing a filter. Fiber variability will be discussed
further in section 1.6 with filter design theory.
Figure 1.4 Meltblown fiber substrate sample produced by Hills Inc. at 2,000x magnification showing variability in fiber size
15
1.3.3 Bi-component
Bi-component fibers, in general, are melt spun fibers and have two polymer phases in the
cross-section13. These fibers were introduced by DuPont in the 1960s with a side-by-side
nylon hosiery yarn, called cantrese. This unique fiber was able to form a highly coiled elastic
fiber upon retraction of the two different nylon polymers. In Asia, specifically Japan, a large
effort in research and development took place in the 1970s37, where the technology was
relatively expensive and due to the complex nature of the spinnerets and spin packs this
technology was not immediately viable for mainstream manufacturing. Further advances in
bi-component technology were seen in 1989 when spin packs made of thin, flat, plates with
holes and grooves were used to route the polymers to conventional melt spinning spinnerets.
This process proved both economically advantageous and flexible for bi-component
manufacturing 37, 38. Figure 5 is a schematic of the bi-component line showing two separate
hoppers and extruders, but a single spin pack which controls the internal fiber shape and
number of fibers by varying the number and type of plates used in conjunction with the
spinneret. Depending on the end use the phases may coexist in order to contribute distinct
properties or as in the case of islands-in-the-sea produce 240-156,000 nano filament fibers as
small as 40 nm in diameter after dissolution of the ―sea‖ component 39.
Bi-component fibers can be fiberized by a dissolvable or splittable process. Fibers produced
via a splittable process include segmented pie or islands-in-the-sea fibers. For example, these
fibers can be carded and then passed under hydro entanglement jet which will both split and
entangle the fibers simultaneously giving the media structural integrity. Islands-in-the-sea
can also be fiberized by dissolving the ‗sea‘ individualizing the remaining ―islands‖. These
16
fibers may demonstrate a variety of cross-sectional geometries. Bicomponent extrusion/
spinning techniques have been used in the textile industry to obtain fibers with physical
response akin to natural fibers e.g., wool which due to its internal structure has a self
crimping nature. In general, side by side bi-components are used to obtain such self bulking
and self crimping properties. Crimping or bulking may occur when the two polymers within
the filament have different strain levels or shrinkage propensity triggered when drawing
tension is removed or when ambient temperatures are met40.
Sheath/core staple fibers consisting of a low melting temperature sheath and higher melting
temperature core for inclusion into nonwovens are today‘s largest commercial use of bi-
component fibers. Sheath/core fibers can also contain a core produced from recycled
material, conductive material or other material that is covered by a sheath that possesses
desired aesthetics or other properties. Micro-denier fibers are used in a variety of industries
but are commonly seen in upholstery for their soft hand and absorbant wipes because of the
large surface area in the interstices in the fabric. Tipped bi-components tend to be used for
special aesthetic properties and bonding. Mixed fiber bi-components can cause unique
aesthetics due to color mixes of fibers. In addition, mixed bi-components can produce bulk
by mixing denier and cross section. Bonding strength can be adjusted to various levels by
mixing homo-polymer filaments with bi-components 39.
Bi-component fibers are advantageous in terms of being spun and processed as larger fibers
but then split or dissolved into non-circular cross sectional geometries such as trilobal,
segmented pie, and other highly complex geometries. As well, they can manufactured at
traditional melt spinning rates. Figure 1.6 shows an extensive list of different variations. In
17
regards to filters, non-circular cross sectional geometries can provide additional bulk without
increasing weight which can lead to better permeability. In addition, the ability to produce
nano-fibers comparable in size to electrospinning but at a much higher production rates is
highly desirable.
Figure 1.5 Schematic of bi-component melt line
18
Figure 1.6 Various bi-component fibers manufactured by Hills Inc.
1.3.3.1 Bi-component electrospinning
It is possible to electrospin two polymer solutions in a side-by-side method to produce bi-
component fibers. This process produces an electrospun mat that possesses properties from
each of the polymeric components. The bi-component electrospinning device includes two
plastic syringes side-by-side which each containing a polymer solution. The target collector
can be any grounded substrate including a rotating cylindrical mesh to obtain filaments
oriented in the extrusion direction, wax paper, Teflon™, and thin polymer film. The two
polymer solutions do not come in physical contact until they reach the end of the spinneret
where the process of fiber formation begins. Gupta and Wilkes27 demonstrated this method
19
by electrospinning a miscible, PVC/Estane®, and an immiscible, PVC/PVDF polymer.
Their studies demonstrated the feasibility of electrospinning bi-component fibers with
diameters in the range of 100nm to a few microns. Their results showed there was some
amount of physical mixing of polymers requiring further investigation into this method.
1.3.4 Flash spinning
Flash spinning is another method of producing ultra fine fibers with filaments varying in size
between 0.1 -0.15 denier with non-circular cross sections 41. This method was accidentally
discovered by DuPont when scientist, Jim White, was exploring explosion behavior of
organic solvents and noticed polyethylene fuzz accumulating on the vent of an experimental
lab42, 43. A polymer is dissolved in a solvent, which is a non-solvent for the polymer at or
below its normal boiling point, and is then extruded from a nozzle at a temperature above
the normal boiling point of the liquid and at higher pressure into a medium of lower
temperature and lower pressure. This flash spinning causes the liquid to vaporize and cool
forming a plexifilamentary film-fibril strand of the polymer. The cobweb network of fibers is
made into a sheet since spinning is too fast to wind. Tyvek™, a product of this method, is
an excellent barrier material which is difficult to tear but easy to cut42, 43. Traditional flash
spinning solvents, such as ethylene chloride and fluorocarbon are said to deplete the ozone
layer making this method of submicron fibers unattractive to many industries. However,
DuPont has made claims in advancements for more environmentally friendly solvents 44 41, 45.
1.4 Filter Manufacturing
Although the definition of a nonwoven fabrics varies, nonwoven filter media has been
described as ‗random fibrous web, formed by either mechanical, wet or air laid means and
20
having interconnecting open area through the cross-section and able to remove a percentage
of particulate from liquid or gaseous fluids streams flowing through it‘46. Nonwoven
production methods are relatively simple and cost effective when compared to other
methods of fabric formation47. Nonwoven materials are commonly used for filters based on
their ability to form stable structures even when they have very low solid volume fractions
(SVFs) which allows for high permeability48. Although nonwovens are not the only method
for high efficiency filtration, they are common because their characteristics and capabilities
can be altered by web construction methods.
1.4.1 Direct Spun Filters
Meltblown and spun bond processes are direct ways to manufacture nonwovens and create
webs in one continuous step via polymers which are melt extruded through a spinneret.
However, the meltblown process forms finer filaments due to a high velocity heated air
stream that converges on the fibers near the die tip13. It is common to see meltblown/spun
bound composite filters since the finer filaments of the meltblown layers provide high
efficiency qualities but are too soft and weak to be used alone. These direct spun filters are
popular due their low cost single step process. Electrospinning is another method of direct
spun filters but as previously stated have low production rates and are commonly formed on
a substrate because of their limited mechanical properties.
1.4.2 Pre-spun Fiber Filters
Nonwovens made from pre-spun fibers require additional steps for final product. In this
method, staple fibers are carded and formed into a web by any of the following processes,
needlepunch, wet laid, chemically or thermally bonded or post-dry46. Used alone or in
21
conjunction, these processes can be adjusted in order to create materials of varying basis
weight (grams per meter2), denier, and thickness, which in turn can impact the permeability
of the end product.
Needle punching, a method which uses barbed needles to entangle and mechanically
interlock the web, is common with baghouse and cartridge filters but due to its high loft not
commonly found in HEPA or ULPA filtration. The wet laid technique, also used in paper
making, is the prominent method for making glass fiber media. The wet lay technique can be
divided into two main segments, the wet end and the dry end. The wet end involves mixing
fibers with water to create a slurry which is then filtered and formed into a wet nonwoven
sheet. The dry end involves the driving out of the water by means of mechanical or thermal
action leaving a uniformly dispersed web. In the final stage of nonwoven fabrication dry or
wet bonding gives the filter media structural integrity 13.
The first step of the wet laid process is web formation. This starts with a combination of
fibers and ingredients which is called the furnish. This furnish is added to water to make a
slurry which then is fed to a wet lay machine. The slurry may encounter various processing
steps, such as refining or fibrillation of the fibers to help them bond better or clean them of
contaminants. Once the wet lay sheet is formed and before it is wound, the water must be
removed and the sheet dried. Mechanical wet presses and suctions are used to remove water
but are not the preferred method for filter media since this action makes the sheets too
dense. The remaining water is removed via thermal drying, such as steam heated can dryers,
hot air dryers, infrared instruments, and microwave heating. The hot air method is preferred
for filters since it allows bulkier nonwovens to be produced. Finally, the wet laid nonwovens
22
are bonded in the wet or dry end or both. At the wet end, a bonding agent can be added to
the slurry before the web is formed, this is referred to as beater addition or beater-ad.
Adding low melting point polymer fibers makes the thermal bonding option favorable,
although, dry end heat is still required to soften/melt thermoplastic binding fibers. Thermal
bonding that makes use of low melting thermoplastic fibers enhances web integrity.
Additionally, the percentage of bonding fibers used will ultimately affect the permeability
and structural properties of the filter. Bonding efficiency is determined by calendar roll
design, temperature, and applied pressure.
Dry end bonding mostly occurs by chemical treatment. Resin application to the web is
common after the web is formed. This process may require some form of wet end bonding
mechanism in addition to hold the web together as the water is removed. In order to alter
the mechanical structure and surface of the web, most wet laid media are subjected to a
converting process. This may include corrugating, slitting and rewinding, creping, die cutting,
pleating, bag making, and sheet cutting13.
Although not as cost or time effective as direct spun filters, pre-spun fiber filters provide
flexibility in filter composition. Specifically, filters produced with staple fibers and carding
techniques allow for the easy blending of fibers. As such, these fiber blends can vary in
denier, polymer, fiber geometry, and ratio amounts, whereas direct spun fibers only have the
ability to vary from layer to layer.
23
1.5 Impact of Fiber Cross Sectional Geometry
Natural fibers do not, in general, have a round or circular section. Silk, for example, is
triangular with round edges in shape providing a high luster4950. Round fibers are still the
most common synthetic fiber shape, although, non-round fibers are becoming more popular
due to their ability to effect fabric surface characteristics 51.
Pertaining to filtration, the use of round glass fibers, typically in the one micron range, is the
HEPA filtration industry standard26,52. However, recently the use of synthetic fibers with
complex geometries is receiving some attention in the field of filtration. For example, trilobal
fibers have been used in needlefelt filter media for their increased surface area. Different
fiber shapes offer different packing configurations which can provide unique air flow paths13.
Rohrback et al53 used a multilobed cross-section fiber to assist with particle and molecular
filtration. In Rohrback‘s case, a reactive reagent, preferably an acid or base in either liquid or
solid form, was placed within the longitudinal slots of the fibers and as contaminants in the
air came in contact with the reagent, they reacted and become fixed to the substrate. The
reagent may be reactive with contaminants, by any known mechanism, such as an acid-base
reaction to form ionic bonds, an oxidation-reduction reaction, and other organic and
inorganic reaction mechanisms which form covalent bonds, hydrogen bonds, coordination
compounds, or complex compounds. The reagent maybe used alone or in combination with
solid adsorptive particles in the slots of the fibers. In addition, a numerical study conducted
at Cornell University with triple Y fibers demonstrated low packing density and relatively
large area to volume ratio which helps with particle collection by Brownian motion54, figure
1.7.
24
Figure 1.7 Geometry of Y and triple Y fibers from Cornell University study on collection efficiency: a) Y fiber b) Y fiber in reversed orientation c) triple Y fiber and d) triple Y fiber in
reversed orientation54
Eastman Chemical Company introduced the 4DG™ fiber in 1997, figure 1.855. In the year
2000 the patent portfolio and technology was donated to Clemson University. 4DG™ is
polyester based shaped fiber and is one member of the Capillary Channel Polymer™ (CCP)
fibers. CCP™ technology is essentially a fiber platform wherein the fibers have higher
surface area compared to round fibers and there are many unique fiber geometries under this
umbrella. The 4DG™ geometry involves a deep groove fiber with 8 surface channels and a
surface area that measures 2.3 – 2.8 times higher than a standard round fiber of the same
denier26.
25
Due to their unique and complex cross-sectional geometry as well as increased surface area,
CCP™ fibers have been investigated for use in a number of products and services. These
fibers were used in tobacco smoke filters since their deep grooved channels resulted in an
increased surface area that allowed higher efficient delivery of tobacco smoke modifying
agents, such as flavorants, to users. 56. Coats, gloves, boats and shoes made with CCP™
fibers showed better insulation since the shape retains the fabric thickness at high pressure
which translates into decreased heat transfer57. As well, this shape demonstrated superior
insulation performance when it was investigated for thermal conductivity performance for
use in space suits58. CCP™ fibers have shown to have twice the fluid adsorption and
transport capacities for oil compared to round and as much show potential for
environmental clean up59. This is of particular current interest in view of the recent
catastrophic ocean oil spill. In addition, the grooved geometry of the fibers could allow for
particle capture, making them a suitable candidate for filtration research26.
Figure 1.8 Cross-sectional image of CCP™ fiber
26
1.6 HEPA and ULPA Filter Design
The crowning goal of filtration is to capture the highest amount of undesirable particulates
while maintaining minimal operating pressure drop and ultimately extending the lifetime of
the filter. Choosing the correct fiber type, fiber diameter, and geometry, in addition to the
correct density and nonwoven process are just a few of the physical parameters that will play
a part in the filter design. In summary, in response to the growing need for more efficient
filters with a longer life expectancy, it is necessary to look at non-traditional fibers and filter
construction for improvement.
1.6.1 Filter Manufacturing Process
Lydall, one of the largest manufactures in North America of HEPA and ULPA filters makes
a large percentage of their air filters with glass fibers, via a wet laid process. The filter is
formed in sheets via the wet laid process, this technique is also used in the papermaking
industry35, 60. Maximizing space and surface area is key to producing a high quality filtration
system. Commonly, HEPA and ULPA filters are pleated in order to provide a higher surface
area for particle collection. 13, 60.
1.6.2 Pre-filters
In order to provide enough surface area and depth to collect the particulates, HEPA filters
are composed of compressed layers of fibers laid either via air or wet laid processes. These
layers can be composed of fibers varying in diameter and type. Since the HEPA‘s filters
overall goal is to collect 99.97% or more of particles 0.3 micron in size, particles larger in
size collect quickly on the surface of the filter causing it to blind off or surface load. This
issue is common to all filtration systems61. One method of resolving this problem is to
27
provide a pre-filter. Pre – HEPA filters can be defined as having efficiencies between 90-
99.97%35 are selected, sized, and installed to maximize the life of the final HEPA filter. For
most applications, the pre-filter is not attached to the HEPA filter so it can be changed out
more often. Since CCP™ fibers provide increased surface area as compared to round; they
have potential of making an excellent pre-filter by providing substantial surface area where
larger particles would collect first before entering the depths of the submicron and nano-
fiber internal layers.
1.6.3 Fiber Size
One of the key areas that fiber size plays a major role is in air permeability and pore size.
Achieving high air permeability to ease pressure difference, while maintaining minimum pore
size to maximum particle collection is ideal, unfortunately here an inverse relationship exists
regarding fiber diameter. Minimizing fiber diameter is essential to improving collection
efficiency and as discussed in Section 1.3, modified meltblown, splittable and dissolvable
bicomponents, and electrospinning are the three main methods of creating ultra fine fibers
but these fine fibers also increase differential pressure across the filter media13.
Fiber filaments must be small enough to create pore sizes that are adequate in stopping
particles, especially 0.3 μm in size, from passing through. Particles 0.3 µm in size are
considered the most penetrating particles size (MPPS) in HEPA and ULPA filtration and
will be discussed in further detail in section 1.8.1.
1.6.4 Fiber Type
Selection of fiber type, such as polymer, glass, or natural fiber, plays a part in filter design. As
discussed, glass fibers historically have played the major role in HEPA filtration because they
28
tend to be inexpensive and have small cross-sectional diameters. Unfortunately, they cannot
be electrostatically charged and are short in length so are unable to be processed through
standard carding nonwoven equipment easily36.
Due to their ability to be functionalized polymer fibers are getting more recognition as a key
contributor in filter design. Polymer fibers can be electrostatically charged, melt spun into
varying shape, which in turn can affect surface area, and the thermal properties such as
melting point and glass transition (tg) temperatures can be tailored via polymer selection and
processing. In addition, strength properties such as tenacity and modulus are relatively high
in polymer fibers and hence they are frequently used in filtration systems.
Although natural fibers such as wood pulp, vegetable, cotton fibers are used in filtration,
they are not found in HEPA and ULPA filters because of their large diameter. Depending
on what the requirements of the final filter are, glass or polymer fibers or combinations of
both are typically used to provide the desired filtration characteristics.
1.6.5 Pore classification
There are three different classifications of pores in a filter, through, blind, and closed pores,
as shown in figure 1.8. Through pores allow flow from one side of the media to the other.
Blind pores terminate within the filter and do not allow air flow but are capable of collecting
particles. On the other hand, closed pores do not participate at all in filtration since they are
completely inaccessible62. Ideally, a filter media composed of through pores for increased air
permeability and blind pores for collection of particulates would be most beneficial for a
filter media. The unique geometry of CCP™ fibers gives them the potential to create a
29
varying tortuous path which could aid in collection of particles while providing through
pores for increased air flow.
Figure 1.9 Different pore types in filtration media
1.6.6 Filter Design with CCP Fibers
As previously stated, an ideal filter would have high collection efficiency with a low pressure
drop. Vaughn26 and Phillips56 spoke about the use of CCP™ fibers in filtration. Vaughn
spoke in regards to the fibers ability to provide increased surface area for collection of
undesirable particles all the while maximizing space available for filtration. He suggests eddy
currents will preferentially deposit particles within the grooves first which will postpone the
blinding off of the pores ultimately extending the life of the filter26. Another advantage of
using CCP™ fibers for filtration is their ability to provide bulk or thickness to the filter
30
without adding weight. This research will examine and expand upon this theory since no
analytical testing has verified these hypotheses.
1.6.7 Composite Filters Composed of Traditional and Non-traditional Fibers
One method of improving HEPA and ULPA filtration is through composite media. It has
been stated that a combination of coarse and fine fibers have higher efficiency and lower
pressure drop than a filter made exclusively of fine fiber63. Podgorski et al64 also investigated
the combination of a nanofibrous layer used in conjunction with a microfibrous support
layer and found the quality of filter much higher than media composed solely of microfibers.
Also noted in this research was the viability of modified melt-blown technique as a filtration
media production method due to its advantages in production speed and fiber size. The
thought and focus of this research is to build on the established principles of HEPA
filtrations but improve and manipulate the construction and building blocks to make its
performance even better.
It has already been established and discussed that metlbown and wet laid nonwovens are a
standard for use in HEPA and ULPA filtration since when constructed properly they meet
the required filtration efficiencies of 99.97% and 99.999%, respectively. One can imagine
with that high of a percentage of particle extraction from the environment how expeditious
filter clogging can occur. No HEPA or ULPA filter can be used indefinitely since it is only a
matter of time before particles collect and energy consumption increases. Through the
correct engineering and design of composite layers, the ultimate demise of the filter may be
extended.
31
1.7 Mechanisms of Filtration
The general term filtration covers a wide spectrum of definitions but when speaking of
HEPA and ULPA filtration the methods are defined for a system of depth filtration where
particles are collected not on the surface of the media but within and are ruled by the
following mechanisms13, and represented as a schematic in figure 1.9:
1. Inertial impaction occurs when the particle inertia is so high that it has sufficient
momentum to break away from air streamlines and impact the fiber.
2. Interception occurs when a particle does not have sufficient inertia to break away from
the streamline, however comes close enough to the fiber so that natural forces will
attach the particle to the fiber.
3. Diffusion is based on Brownian (zig zag) motion of very small particles (<0.5µm).
This random and probabilistic motion will cause a particle to vary from the
streamline and possibly engage a fiber.
4. Electrostatic attraction is based on an electric or electrostatic charge on the particle
and/or fiber that will force the particle to be diverted from the streamline and
attracted to the fiber.
Particles that are 0.3 µm in size or larger are mostly captured by mechanisms of inertial and
interception in fibers based filters65. Generally, fiber diameter is inversely related to the flow
resistance of the filter media. The rule states the smaller the diameter of the fiber the greater
the resistance13. On the other hand, the filter efficiency of a fabric made from a finer fiber
will increase66. The effect of slip flow must also be noted, for if this theory is valid, then
fiber diameters of a small enough proportion, under 500 nm, will have the benefit of
32
increased collection efficiency all the while maintaining or decreasing pressure drop as
discussed in section 1.2.1.
These mechanisms are based on round cross-sectional fibers. This research will examine and
try to understand if the mechanisms are the same or different for non-round cross-sectional
shaped fibers, as well as if the theory of slip-flow effect occurs for round fibers with
diameters measuring 0.5 µm or smaller.
Figure 1.10 Key mechanisms of filtration during HEPA and ULPA filtration
33
1.7.1 Most Penetrating Particle Size (MPPS)
Particles that are the most difficult to capture are referred to as the most penetrating particle
size (MPPS). Smaller particles break away from the air stream, move by Brownian motion
and collect and hold to a fiber surface via Van der Waals forces67. While larger particles,
having more momentum, are caught by direct interception or inertia mechanism. There is a
particle size range, around 0.04 – 0.4 micron, that is too large for diffusion but not large
enough to be captured by inertial mechanism. For HEPA filtration, the MPPS is usually 0.3
micron13. Figure 1.10 is a schematic of this phenomenon of efficiency versus particle size.
Figure 1.11 Schematic of collection efficiency vs. particle size of most penetrating particle
size (MPPS)13
MPPS can also change with changes in air flow velocity and filter type. As velocity and filter
solidity increases, MPPS decreases. In addition, as velocity decreases, collection efficiency
0
0.2
0.4
0.6
0.8
1
0 0.2 0.4 0.6 0.8 1 1.2
Fra
cti
on
al C
oll
ecti
on
Eff
icie
ncy
Particle Size (microns)
34
increases since particles have more time to be captured by Brownian motion67. European
standards for HEPA and ULPA filters are based on filtration efficiency of the MPPS13
Based on these established industry test standard of MPPS, collection efficiency testing for
the purposes of this research will also center on MPPS and achieving a 99.97% or better
efficiency against 0.3 micron particles for HEPA filtration and 99.999% or better against 0.1-
0.2 micron particles for ULPA filtration.
1.8 Aim of Research
Submicron and nano-scale fiber manufacturing processes play a part in filter design and will
be a considerable focus in this research. Although a proven and effective method of nano-
fiber production this research will not investigate electrospinning. Instead, it will examine
other production methods which are quicker and may provide fibers comparable in size. Bi-
component fibers will be looked at since they provide a wide variety of avenues to obtain
shape and size variations at the same manufacturing production level of melt spun fibers.
Specifically, bi-components will be examined as a method to produce filter media via a wet-
laid process made with fibers comparable in size to those produced from indolent
electrospinning methods.
CCP™ fibers will be examined for how their unique shape and attributes contribute to
filtration. As discussed, these fibers have shown to have qualities that exceed that of round
fibers. Used alone or in conjunction with other fibers in a nonwoven filter media, CCP™
fibers may provide the ability to improve filtration. A successful filter design is dependent
on many factors. Designing a filter composed of CCP™ fibers of varying size with their
35
increased surface area and grooves should help not only provide additional areas for the
contaminants to adhere to but also provide varying tortuous paths which can aid in the
reduction of pressure drop across the filter. The CCP™ fibers could allow for a combination
of both through and blind pores to allow air flow and particle collection respectively. To
date, filter research has not examined many areas of manipulating fiber geometry to help
improve performance26.
This research also focuses on the theories of slip flow versus non-slip flow. Examining the
ability for fibers fewer than 0.5 µm in diameter to improve collection efficiency while
reducing or maintaining differential pressure would be a great breakthrough in quality HEPA
filters. There is speculation in current findings with the slip flow theory since electrospun
fibers not only vary in diameter but also require a substrate with fibers having diameters
larger than 500 nm which would impact the theory. In addition, this theory only works for
clean filters, as particles begin to build quickly on nano-fibers, the theory may be incorrect
over any reasonable time scale of filter use68. Through the use of bi-component fibers it is
possible to test these theories more thoroughly. Bi-component fibers, such as the 156,000
islands-in-the-sea fibers provide a consistent supply of fibers with diameters less than 500
nm. In addition, they can be processed via a wet laid system which is quicker than
electrospinning making them more viable in industry.
This research reexamines traditional HEPA and ULPA filters to improve upon the
governing parameters of filtration, reducing pressure drop and increasing collection
efficiency by examining fiber geometry and filter construction. Specifically, if CCP™ fibers
preferentially collect particles deep within their grooves then collection efficiency may be
36
increased all the while maintaining or reducing pressure drop. As well, if modified
meltblown and bi-component fibers can be processed into submicron and nano-fibers then
slip flow theory can be tested. The scope of this research is to examine current engineering
and science of HEPA and ULPA filter media and develop a cost effective, sustainable
method of increasing collection efficiency, and maintaining or reducing pressure drop.
37
Chapter 2
EXPERIMENTAL
2.1 Materials
Table 2.1. List of materials used during experiments
Chemical/Material Supplier Description
.84 IV PET Wellman Small white pellets Molecular Formula
(C10H8O4)n
Exceval Kuraray Water soluble polymer
Polyvinyl alcohol
ESPET Kuraray Easy Soluble Polyester
WSPET Eastone Water soluble Polyester
Molecular Formula
Low Viscosity Embedding Media Electron Microscopy Sciences Embedding resin
Triton x-114 Aldrich Surfactant
Sodium Hydroxide Fisher Scientific Molecular Formula NaOH
Butyl Alcohol Normal Mallinckrodt Chemical
Plasticizer Molecular Formula
CH3(CH)2CH2OH
Benzyl Alcohol Aldrich
Plasticizer Molecular Formula
C6H6(CH)2OH
Krazy Glue® CVS drugstore Instant Krazy glue gel
Table 2.2 is a list of melt spun, bi-component, and melt blown fibers and fabrics which were
investigated in this dissertation for use in HEPA and ULPA filtration. In addition to these
nonwovens, (M98) filter media will be used as a benchmark for comparison and testing since
this filter media passes all HEPA industry standards. Funding for this research was provided
38
by Naval Surface Warfare Center Dahlgren (NSWCDD) which currently uses the M98
HEPA filter media.
Melt spun CCP™ fibers have listed denier which is determined by winding 9,000 meters of
fibers and measuring the mass. Denier is defined as mass (in grams) of 9,000 meters of fiber.
Table 2.2 List of melt spun, bi-component, and melt blown fibers investigated for use as HEPA and ULPA filtration media as well as M98 filter media used as a benchmark material
Melt Spun Cross Section Polymer Denier
Capillary Channel Polymer PET 3 Capillary Channel Polymer PET 6 Capillary Channel Polymer PET 15 Capillary Channel Polymer PET 30
Bicomponent Cross Section Polymer (%) Separation Process
Capillary Channel Polymer Islands Polypropylene 19 Dissolvable
Round Sea Polylactic Acid 81
Capillary Channel Polymer Islands Nylon 6 19 Dissolvable
Round Sea EXCEVAL (PVA) 81
Capillary Channel Polymer Islands Nylon 6 19 Dissolvable
Round Sea ESPET 81
Snowflake Islands PET 40 Dissolvable
Round Sea WSPET 60
Snowflake Islands Nylon 6 25 Dissolvable
Round Sea EXCEVAL (PVA) 75
39
Snowflake Islands Nylon 6 20 Dissolvable
Round Sea ESPET 80
156K Islands Nylon 6 10 Dissolvable
Round Sea ESPET 90
Meltblown Sample ID Polymer Basis Weight (g/m2) PE Scrim
111808-01 PP 10.5 No
111808-02 PP 5.1 No
111808-03 PP 15.5 Yes
111808-04 PP 11.8 Yes
111808-05 PP 10.3 Yes
111808-06 PP 11.2 No
111808-07 PP 13.0 Yes
111808-08 PP 11.1 Yes
111808-09 PP 10.9 Yes
111808-10 PP 13.2 Yes
111808-11 PP 11.3 Yes
111808-12 PP 11.6 Yes
Clemson MB7 PP 14.0 Yes
99 031008-01 PP 7.0 No
Lydall Filter Media Sample ID Polymer (%) Basis Weight (g/m2)
3333B Glass 77 69.7
Acrylic 3
PET 20
2.2 Fiber Extrusion
Hills Research and Development Pilot Extruder, Model REM-3P-24 was used to extrude
CCP™ PET fibers, 6, 15, 20, and 30 dpf in size. The extruder operates with a 1.0‖ screw
equipped with four heating zones, with a final extrusion temperature of 285°C. Pump speed
varied between 10.2, 14.1, and 17, revolutions per minute (rpm), depending on denier, while
40
the pump size was .584 cubic centimeter per revolutions. The spinneret was CCP™ shaped
with 16 holes. Prior to extrusion, .84 IV polymer chips were dried in a Maguire Low
Pressure drier for 30 minute at 150°C. All other fibers used for this research were
manufactured by Hills, Inc.
2.3 Bi-component Dissolution
Continuous bi-component filaments were cut to an approximate length of 0.5 inches for
splitting by solvation and agitation. To accomplish this, .1-1.0 grams of fibers were placed in
250 ml jars filled with water plus sodium hydroxide, sodium sulphate, or butyl or benzyl
alcohol as solvents. The solution was heated up to 65°C for as long as 7 days. A VWR
Model 150HT sonicator was set for 30 minute cycles unless otherwise noted. All solutions
were made with de-ionized (DI) water. Fibers were rinsed under tap water between solvent
changes.
2.4 Nonwoven Manufacturing
2.4.1 Dry Laid Filter
Continuous CCP ™ filaments were hand cut by scissors to approximately 2 inch staple
length fibers. The samples were manually opened by hand before being placed into a CMC
Rando for further mechanical opening and mixing. In addition to the shaped fibers, 20% (by
weight) of a low melting point 4 dpf polyester (PET) binder fiber was blended into the
mixture to serve as a binding agent during the thermal consolidation of the final nonwoven
webs. Nonwovens were produced by running the fibers through a 20" Proctor & Schwartz
Roll-Top Card followed by immediate point bonding on a 20" Benz Thermal Bonding
41
Calendar which was set to 130°C and 30 kilopascal per centimeter. When producing a
multilayered nonwoven sample, the single layered nonwovens were processed through the
point bonder again set at 120°C and 20 kilopascal per centimeter until desired basis weight
was achieved.
2.4.2 Wet Laid Filter
Dissolvable bi-component fibers processed via a wet-laid method were prepared with a
Buchner funnel and flask as demonstrated in figure 2.1. The fibers were hand cut to between
0.125 and 0.5 inch and laid in the Buchner funnel on top of filter media. The solvent, after
being heated, was poured over the fibers and the solution was removed by alternating
between a vacuum pump and gravity. Details of solvents used are discussed in section 3.6.
The vacuum pump expedited the removal of the solution while gravity alone allowed the
fibers to spend more time in the solvent. A vacuum pump motor model: S55NXMLD-67,
H.P. 1/3, RPM 1725, was used to remove the solution from the fibers in the Buchner
funnel.
Figure 2.1 Schematic of Buchner funnel used for wet-laid filter media production
42
2.5 Characterization of Fibers
All fibers were examined by optical or electron microscopy imaging for characterization of
size, size distribution, and shape. When examining CCP ™ fibers, shape integrity, fiber
length, and groove size was examined.
2.5.1 Fiber Sample Preparation for Optical Microscopy
CCP ™ fibers, 3 dpf in size and larger, were examined under the optical microscope for
shape integrity. Cross sectional images of the fibers were prepared by microtoning the
sample with a Hills microtone, figure 2.2.
Figure 2.2 Hills microtone used during cross-sectional fiber preparation for optical microscopy
2.5.2 Fiber Sample Preparation for Electron Microscopy
For fibers smaller in size than 3 dpf, it was necessary to use an electron microscope in order
to get adequate magnification for shape analysis. To view cross-sectional images of fibers for
electron microscopy, samples had to be embedded in resin and microtomed first. Fibers
were embedded in a Beem Embedding Capsule as shown in figure 2.3. Store bought Crazy
Glue was used to fill the hole at the bottom of the capsule. A 10 ml syringe with a 20 gauge
43
needle was used to fill the capsule with a low viscosity embedding media. The resin was
prepared ahead of time using the ‗firm‘ recipe. Once the capsules were filled, they were
placed in a Sheldon Manufacturing Inc. oven overnight at 70°C. Once cured, the samples
were removed from the embedding cap.
Figure 2.3 Embedding cap with stand used for cross-sectional fiber preparation
2.5.2.1 Microtoming of Fiber Samples
A Reichert-Jung Ultracut E Microtome was used to slice away sections of resin and fiber
around 200 nm thick. A glass blade was used repeatedly to remove sections of the resin until
a smooth surface exposing the fibers was present. Once the fibers embedded in the resin had
a smooth exterior, they were ready for electron microscopy imaging.
44
2.5.3 Optical Microscopy
Optical images were obtained using a Vistavision optical microscope equipped with
ProgRes® CapturePro 2.0 software. This software allowed for capturing of images as well as
measuring linear distance.
2.5.4 Electron Microscopy
Scanning electron microscopy (SEM) imaging work was performed on an S-3400N or
FESEM Hitachi S4800. SEMS provide topography images by emitting an electron beam
towards a sample and depending on how the electrons interact with the sample‘s atoms
provides information about its surface. The S-3400N is designed for conventional and
variable pressure microscopy. It is equipped with an Oxford INCA EDS, WDS, EBSD and
built-in four quadrant solid-state backscatter detector. Also, it has SEM imaging capability in
variable pressure. The S4800 offers field emission for high resolution microscopy. It is
equipped with an Oxford INCA Energy 200 EDS and a GW Electronics Centaurus
backscatter detector. Each individual image, an example is seen in figure 2.4, is marked with
scope, beam current (kV), working distance, magnification, detector type, vacuum setting
(Pa), and time information. Fibers or fibers processed into nonwovens were sputter coated
with platinum for two minutes prior to imaging unless viewed in an embedding cap. Cross
sectional imaging processed via an embedding cap were required to be performed on the S-
3400N due its ability to function under variable pressure.
45
Figure 2.4 Example of SEM image with scope, beam current (kV), working distance, magnification, detector type, vacuum setting (Pa), and time information
2.5.4.1 Micrographs Analysis
Universal Desktop Ruler v3.0.1211 from AVPSoft.com was used to analyze fiber length,
fiber diameter and also determine surface area. This software allows the user to trace
individual fibers in the micrograph with connected line segments. The software is calibrated
against the scale bar from the micrograph. Measurements were taken for all filaments visible
in an image and the values for perimeter and area were recorded and an average value was
reported. From these calculations, surface area was calculated by multiplying length by
perimeter. Sample length in meters was determined by dividing the sample mass by the
denier per filament and multiplying by 9,000.
46
2.6 Filter Characterization
2.6.1 Deconstruction of M98 Filter Media
In order to evaluate M98 filter media for benchmarking analysis, the media was
deconstructed. Exterior and interior layers were examined for fiber size distribution, polymer
composition, as well as overall filter construction. This was accomplished through
examination of SEM images and FT-IR spectra both pre and post TGA runs. The paper-like
media was deconstructed by peeling the layers away from each other by tweezers. Interior
layers were considered any layer that was exposed after peeling off an exterior layer since
precise delamination of the filter media was not possible.
2.6.2 Capillary Flow Porometer
Capillary Flow Porometer (CFP), model CFP-1100-AEXS, manufactured by Porous
Materials, Inc., was used to analysis the nonwoven fabrics. The CFP is a nondestructive
method capable of providing smallest detected pore diameter, mean flow pore diameter,
bubble point pore diameter, diameter at maximum pore size distribution, and differential
pressure. The change of flow rate (of compressed gas) with pressure is recorded through the
samples. These measurements were used for preliminary test results in terms of analyzing
overall pore size and differential pressure
Nonwovens were hand cut by scissors into 4 cm by 3 cm samples. Individual samples were
placed between the bottom of the sample chamber and the chamber insert secured by O
rings in order to verify no air leakage, figure 2.5. Nonwoven samples tested required a
minimum of two layers since one layer had too high air permeability and the instrument was
unable to detect change of pressure which allows it to record data. All samples were tested
47
using a dry up, wet up test using Galwick wetting agent. The differential pressures and gas
flow rates through dry and wet samples are measured. In the dry sample, the flow rate
increases with increase in pressure. In case of the wet sample, initially there is no flow
because all the pores are filled with the liquid. At a certain pressure the gas empties the
largest pore and gas flow starts through the wet sample. With further increase in pressure
smaller pores are emptied and the flow rate increases until all the pores are empty and the
flow rate through the wet sample is the same as that through the dry sample. Smallest
detected pore diameter, mean flow pore diameter, bubble point pore diameter, diameter at
maximum pore size distribution, and differential pressure are measured and recorded by the
CFP software package.
Figure 2.5 Schematic of Capillary Flow Porometer
48
2.6.3 TSI Fractional Efficiency Testing
All fractional efficiency testing was conducted at the Naval Surface Warfare Center in
Dahlgren, Va. The TSI fractional efficiency tester, model TSI 3160 is capable of testing
efficiency as well as pressure drop across filtration media. The instrument operates by
inserting a filter and passing an aerosol through the filter. Two light-scattering laser
photometers simultaneously measure the upstream and downstream aerosol concentration
levels. The particle penetration value is determined from the ratio of these two readings. By
using two laser photometers, instead of one, measurement accuracy and throughput is
increased. The 3160 model is able to test media up to 99.999999%. Particle ranging from
0.015 to 0.400 µm in size were used for testing.
2.6.4 Salt Deposition Study
Salt deposition studies were conducted by John Larzelere at the Naval Surface Warfare
Center in Dahlgren, Va. A PALAS MFP Filter Test System, model MFP 3000, was used to
conduct salt deposition studies on CCP ™ fibers. This modular instrument is capable of
testing for fractional efficiency, loading capacity/lifetime, and pressure drop and is equipped
with a light-scattering spectrometer. The MFP 3000 works in the suction modus which
ensures an especially even formation of the dust cake also at high inflow velocities. The
light-scattering spectrometer provides a reliable determination of the aerosol concentration
and the particle size is ensured. For purposes of this research, it was only used to spray the
shaped fibers with salt particles, .05, 0.1, and 0.4 μm in size in order to analyze loading
tendencies. The samples were examined with SEM microscopy imaging after particle
loading.
49
2.6.5 Fourier Transform Infrared (FTIR) Spectroscopy
Fourier transform infrared (FTIR) spectroscopy allows researchers to identify composition
of materials by using radiation from the mid-infrared spectral region (4,000 to 200 cm-1) to
excite bonds within molecules. The energy beam is analyzed for its absorption by the
molecule and a spectrum is formed after a Michelson Interferometer converts the energy
beam to an interferogram which is then converted by a fourier transform to an IR spectrum.
Samples were analyzed with a Thermo-Nicolet Magna 550 FTIR with an ATR accessory and
Omnic software for composition verification. A small sample was cut by scissors and placed
into the FTIR diamond window. A background spectrum as well as a sample spectrum was
collected. ATR corrections were made to the spectra in order to make them comparable to
transmission FRIR spectra. Once these corrections were made, the sample was than analyzed
for characteristic peaks which are compared to a library data base.
2.6.6 Thermogravimetric analysis (TGA)
Thermogravimatric analysis (TGA) was used in conjunction with FTIR for analysis of bi-
component fibers and filter media. TGA measures changes of weight of a sample with
change of temperature which can help determine mass loss over a specific temperature range
and final decomposition temperature. A Hi-Res TGA 2950 Thermogravimetric Analyzer was
used to obtain mass loss characteristics of the samples by increasing the temperature of the
sample until entire polymer components were burned off leaving only the glass fibers. A
sample close to 5 mg in weight was cut and placed into a clean platinum pan. The instrument
is purged with nitrogen after the sample is loaded and the precise sample weight was
measured. The instrument was slowly heated to 600°C, in order to ensure only glass fibers
50
remain, and sample weight change is recorded. Once the final weight is measured, the
quantitative composition of glass to polymer fibers can be determined. FTIR was rerun on
this sample to verify glass characteristic peaks are only visible and no polymer material is left.
Microscopy imaging of the remaining glass fiber was analyzed for fiber size distribution.
2.6.7 Differential Scanning Calorimetry (DSC)
Differential scanning calorimetry (DSC) is a technique used to evaluate thermal properties of
polymer samples. Through this process chemical reactions and physical changes are
monitored with change in temperature. The heat flow to the sample is plotted versus the
temperature which gives a differential thermogram that identifies phase transitions of the
polymer. Bi-component fibers were examined using DSC techniques in order to determine
information regarding their crystallinity, glass transition (Tg) and melting temperature (Tm).
A DSC 2020 Differential Scanning Calorimeter made by TA instruments was used for these
tests. Samples were heated under nitrogen (other inert gases may be used). Small sample
fiber cuts (between 4-5.5 mg) were placed in a DSC pan and chamber. The samples were
heated at a rate of 20°C per minute until 300°C for a minute in an isothermal step. The
samples were then removed from the chamber and immediately quenched on a bar cooled
with nitrogen. These thermograms, (before quench), gives information on the fibers as
received and in their spun state. The samples were reloaded in the DSC chamber and rerun
under the same conditions. The second set of thermograms (after quench) provided
information on the fibers after erasing the thermal history of the samples.
51
2.7 Modeling Simulation
Finite element analysis and Brownian dynamic simulation were used to predict particle
collection and differential pressure across filter media. Fiber and nonwoven specifications,
including polymer type, fiber geometry, denier, thickness, basis weight, and cross sectional
dimensions from experimental samples were simulated in a mathematical model. The Stokes
equation was relied upon as the governing equation for fluid flow analysis. Modeling
simulation studies were conducted by Dr. Christopher Cox and Patrick Buckingham.
52
Chapter 3
RESULTS and DISCUSSION
The rationale for using CCP ™ fibers in HEPA and ULPA filter media was to increase
surface area which in theory should increase collection efficiency by providing additional
particulate collection sites. In addition, particulates should preferentially deposit within the
deep groves and surface channels aiding in postponing the blinding off of the pores leading
to life extension of the filter26. The main focus of this research was to evaluate whether the
CCP™ fiber‘s unique geometry supplied additional benefits in terms of increasing collection
efficiency and lifespan of a HEPA quality filtration system while minimizing the energy
demands caused by increased pressure drop. The proposed mechanism for this improved
filter media performance relies upon the increased surface area at equivalent weight, which
provides for additional particulate collection sites.
In order to appreciate any advancement in filtration efficiency a benchmark was established
for comparisons reasons. M98 media was the standard which was used during the scope of
this research since it meets HEPA industry collection efficiency ratings of 99.97% or better
of particles 0.3 µm in size.
3.1 Benchmarking of M98 Filter Media
3.1.1 SEM imaging of M98 fibers
The M98 filter media meets HEPA standards and was the benchmark for all testing in this
research. Deconstruction of this media was conducted in order to examine the sum of its
parts and thus the media was examined using electron microscopy. The cross section,
53
exterior, and inner layers were examined for fiber size and overall construction. Figures 3.1
and figure 3.2 display the cross section of the filter media. Figure 3.1 is an SEM image at
250x magnification taken at the end of the edge of the media where the fibers are aligned
mostly parallel to each other. Figure 3.2 shows the media at a lower magnification of 150x
where the fibers appear to be laid in sheet-like layers. Figure 3.3 and 3.4 displays the exterior
of the media at 110x and 8,000x magnification respectively. Figure 3.3 shows the overview
of the exterior of the filter with pronounced larger fibers. Figure 3.4 exhibits variations in
fiber diameters as well as evidence of a polymer film (presumably a binder) that bridges the
fibers. Figures 3.5 and 3.6 are images of the inner layer of the filter. Figure 3.5 shows the
interior layer at 60x magnification where the larger fibers appear to be laid in a light mesh
configuration. Figure 3.6 displays the interior layer at 9,000x magnification again showing
variations in fiber diameter. It becomes evident from these images there is range of fiber size
in the M98 filter media from submicron to several microns in diameter.
54
Figure 3.1 SEM image taken at the end of the edge of a cross section of M98 filter media
Figure 3.2 SEM image of cross section of M98 filter media where fibers appear to be in sheet-like layers
55
Figure 3.3 SEM image of exterior layer of M98 filter media at 110x magnification
Figure 3.4 SEM image of exterior layer of M98 filter media at 8,000x magnification showing variability of fiber diameter as well as polymer binder film
56
Figure 3.5 SEM image of inner layer of M98 filter media at 60x magnification where larger fibers appear to be laid in a light netting effect
Figure 3.6 SEM image of interior layer of M98 filter media at 9,000x magnification showing
fibers varying in diameter
57
3.1.2 Fiber size distribution of M98 media
Since fiber size plays such an important role in filtration, fiber size diameters were measured
from SEM images and a fiber size distribution graph was produced, figure 3 7. The
measurements represented in this figure were taken from exterior and interior fibers from
the M98 media. For this work, fibers that are under 100 nm will be considered nano-fibers.
Approximately 44% of the fibers were measured in the 0 .101 – 0.50 µm range and only 5%
of the fibers were in the 0.01-0.10 µm range. While 24% of the fibers were measured in the
0.501-1.01 µm range and 27% were measured in the 1.01 µm and above size range. Keeping
in mind since the measurements were taken from SEM images the sample size is small
compared to the overall filter size. Based on section 1.2.1 discussion of slip flow theory
which applies to fibers less than 0.5 µm in diameter, this would indicate that 49% of the
fibers are within the slip flow range potentially contributing towards collection efficiency.
All the while they are reducing or maintaining pressure drop. Since this research is striving
towards making improvements to the current filtration standards, one suggestion would be
to introduce a larger amount of fibers within the slip flow range. One method for
consistently producing fibers smaller than 500 nm is via bi-component extrusion processes
and meltblown processes at a semi-commercial scale. These methods will be examined in
further detail later in this chapter.
58
Figure 3.7 Fiber size combined distributions of M98 polymer and glass fibers
3.1.3 Glass composition analysis
The M98 media was examined by ATR-IR, pre and post TGA, in order to establish
composition. Initial ATR-IR test indicates characteristic peaks of PET, glass, and possible
acrylate styrene co-polymer. Acrylate styrene co-polymer is commonly used as a binder in
paper and other applications because of its adhesion capabilities to a wide variety of
substrates, as well as toughness, good flexibility, and excellent chemical and water
resistance111. Post-TGA, ATR-IR was rerun and showed only characteristic peaks of glass
confirming no polymer fibers remained. Figure 3.8 is the ATR-IR spectrum which shows
results for both pre and post TGA run. Table 3.1 lists characteristic assignments of
frequencies which would correlate to the glass and PET fibers as well as possible acrylate
styrene.
0%
5%
10%
15%
20%
25%
30%
35%
40%
45%
50%
.01-0.1 .101-0.5 0.501-1.00 1.01-above
Fiber Size in Microns
M98 Fiber Size Distribution
59
TGA was run with a sample starting weight of 4.467mg. The sample was initially heated to
500°C at 20°C per minute in nitrogen. A second TGA cycle was run at 20°C per minute in
nitrogen where the weight leveled off around 550°C in order to ensure all polymer fibers
were burned off in turn leaving only the glass fibers, figure 3.9. The sample weighed 3.410
mg post TGA indicating the glass fibers made up 76.13% of the filter composition.
Figure 3.8 ATR-IR spectra of M98 filter media post and pre TGA showing characteristic glass peaks only post TGA and peaks of PET and glass prior to TGA
60
Table 3.1 Group frequency assignments for glass, PET, and polystyrene69,70
Frequency (cm-1)
Assignment Polymer characteristic peak
1,050 Si-O stretch Glass
1,725 C=O stretch Poly (ethylene terephthalate)
2,969 CH2
asymmetric stretch Poly (ethylene terephthalate)
Figure 3.9 TGA thermogram of M98 filter media
61
SEM images were taken to examine the glass fibers post-TGA. Examples of these images
can be seen in figures 3.10 through 3.12. These images showed there was still variation in
fiber diameter and fiber size. Measurements were taken in order to analyze fiber size
distribution. Figure 3.13 shows the fiber size distribution post-TGA. As with figure 3.7,
which showed fiber size distribution for all M98 fibers, the majority of glass fibers, 41%,
were in the .101-.500 µm size range. Once again these fibers are submicron but not nano-
size. Figure 3.14 shows the fiber size distribution of the M98 filter media exterior and
interior fibers separately, as well as the post-TGA fibers which were imaged as exterior
fibers. Figure 3.15 represents the comparison of M98 fiber size diameters of as received
versus post-TGA. The overall impression of the M98 filter media is that its composition fits
with traditional HEPA media description of being composed of a majority of round glass
fibers with most of the fibers in the submicron range but a low percent (under 10%) being in
the nano-size range.
62
Figure 3.10 M98 filter media composed of only glass fibers after pyrolysis in TGA to remove polymer fibers at 400x magnification
Figure 3.11 M98 filter media composed of only glass fibers after pyrolysis in TGA to remove polymer fibers at 1000x magnification
63
Figure 3.12 M98 filter media composed of only glass fibers after pyrolysis in TGA to remove polymer fibers at 9,000x magnification
Figure 3.13 M98 glass fiber size distributions after pyrolysis in TGA to remove polymer fibers
0.00%
5.00%
10.00%
15.00%
20.00%
25.00%
30.00%
35.00%
40.00%
45.00%
.01-0.1 .101-0.5 0.501-1.00 1.01-above
Fiber Size in Microns
M98 Fiber Size Distribution of Glass Fibers
64
Figure 3.14 Fiber size distribution of M98 filter media pre-TGA and post-TGA
Figure 3.15 Comparison of M98 as received and post-TGA fiber size diameters
0%
10%
20%
30%
40%
50%
60%
70%
.01-0.1 .101-0.5 .501-1.0 1.01 & above
Fiber Size (µm)
M98 Fiber Size Distribution
As Received Exterior
As Received Interior
Post TGA
0
2
4
6
8
10
12
14
16
Fib
er
Dia
mete
r (µ
m)
M98 Fiber Size Distribution
M98 As Recieved
M98 Post TGA
65
3.1.3 M98 collection efficiency
M98 media was evaluated for collection efficiency since this material was the benchmark for
current filtration standards. Figure 3.16 graphically represents the collection efficiency versus
flow rate of the M98 filter media at 0.3 µm particle which is a common particle size when
evaluating HEPA media. The media was determined to meet HEPA standards when tested
at flow rates up to 10.0 L/min but did not meet the standards for flow rates above that limit.
Figure 3.17 shows the collection efficiency versus pressure drop with 0.3 µm particle size.
The collection efficiency decreased with increased pressure drop which correlates with figure
3.16 since the flow rate also increases. Figures 3.16 and 3.17 represent collection efficiencies
and pressure drop, respectively, for 11 separate tests on M98 media at 0.3 µm particle size.
Figure 3.16 Collection efficiency vs. flow rate of M98 filter media at 0.3 µm particle size
99.92
99.93
99.94
99.95
99.96
99.97
99.98
99.99
0 5 10 15 20 25 30 35 40 45
Co
llecti
on
Eff
icie
ny (
%)
Flow Rate (L/min)
M98 Collection Efficiency vs. Flow Rate at 0.3 µmParticle Size
66
Figure 3.17 Collection efficiency vs. pressure drop of M98 filter media at 0.3µ particle size
Collection efficiency was also evaluated at challenge particle sizes at or below 0.3 µm, which
was determined experimentally to be the most penetrating particle size (MPPS) for the media
evaluated during this research. As well, as discussed in section 1.7.1, MPPS can vary with air
velocity. Figure 3.18 shows results of collection efficiency versus flow rate of the media at its
most penetrating particle size (µm). Slower flow rates allowed for higher collection efficiency
since the particles spend more potential residence time within the media allowing for
collection via one of the mechanisms of filtration. The MPPS for all flow rates varied
between 0.100-0.200 µm. At the most penetrating particle size the M98 media did not meet
HEPA collection efficiency standards of 99.97%. At the lowest flow rate of 8.35 L/Min, the
collection efficiency was 99.96%, and at the highest flow rate of 40.02 L/Min the collection
efficiency was at 99.56%.
99.92
99.93
99.94
99.95
99.96
99.97
99.98
99.99
0 5 10 15 20 25 30 35 40
Co
llecti
on
Eff
icie
ncy (%
)
Pressure Drop (mmH2O)
M98 Collection Efficiency vs. Pressure Drop at 0.3 µm Particle Size
67
Figure 3.18 Collection efficiency vs. flow rate of M98 filter media at MPPS (µm)
3.2 Characterization of Capillary Channel Polymer™ Fibers
As discussed in section 3.1, M98 filter media was the standard by which HEPA filtration was
defined for purposes of this research. This media met the 99.97% collection efficiency of
particles 0.3 µm in size at flow rates of 10.0 L/min or lower but like all filters pressure drop
also played a large role in filter standards. The research results supplied in this chapter
examined not only CCP™ fibers used alone in a nonwoven filter media but as well as part of
a composite material. This analysis examined the ability of the CCP™ fibers to provide
advances in collection efficiency all the while maintaining and/or reducing pressure drop.
99.52
99.57
99.62
99.67
99.72
99.77
99.82
99.87
99.92
99.97
0 10 20 30 40 50
Co
llecti
on
Eff
icie
ncy (
%)
Flow Rate (L/min)
M98 Collection Efficiency vs. Flow Rate at Most Penetrating Particle Size (µm)
.126 MPPS
.126 MPPS
.126 MPPS
.200 MPPS
.109 MPPS
.109 MPPS
.109 MPPS
.100 MPPS
.100 MPPS
.100 MPPS
.100 MPPS
68
3.2.1 Shape Integrity of CCP™ Fibers
Filters must be able to collect particulates varying in size, keeping in mind HEPA filters are
commonly evaluated at 0.3 µm13. In practice particulates removed via air filters can measure
within a wide range of size, as demonstrated in figure 3.19. ULPA requirements are for
smaller size particles, generally 0.1-0.2 µm but may vary with face velocity13. CCP™ fibers
with their varying groove sizes must be able to collect particles varying in size as well. In
order to collect the contaminants within their deep grooves, shape integrity was necessary.
Figure 3.19 Relative size of common materials and common air contaminants13
Shaped fibers can experience more distortion of shape than round fibers after melt spinning.
When extruding round fibers, surface tension forces of the polymer aid the fiber in retaining
the round shape of the spinneret hole. With shaped fibers, the surface tension and other
forces can cause the shape to become distorted from the original spinneret design. With
69
shaped fiber it is thought the elongational stress and wall normal stress as it forced through
the spinneret can cause distortion of the shape59. Since maintaining that shape is pivotal for
understanding if and how that geometry can assist in improving filtration, it was necessary to
examine shape integrity by optical and SEM imaging.
CCP™ fibers 3 dpf and larger were viewed via an optical microscope and fibers smaller than
3 dpf were viewed by electron microscope. Random sampling to determine the consistency
of fiber shape in the CCP™ fiber utilized in the research were examined under SEM for
detailed measurements of groove size. CCP™ fibers with proper shape integrity were
defined as having distinct and visible eight legs and grooves. Figure 3.20 is an image of a 30
dpf polypropylene CCP™ fiber where the shape integrity was intact. This image was taken
with an optical microscope. Figure 3.21 is an SEM image of a bi-component EVOH sea and
Nylon 6 island fiber. Shape integrity was also intact although variation in shape between the
two samples is evident. Figure 3.22 is an SEM image of a bi-component PLA/PP fiber
where shape integrity was very poor. All fibers not demonstrating the proper unique
geometry of the CCP™ were excluded from this research. All fibers extruded at Clemson
University or bobbins manufactured by Hills Inc. had cross sectional samples taken from
them and were examined for shape integrity. The bi-component PLA/PP fibers were not
used for any part of this research due to their distorted shape. All other CCP™ fibers listed
Table 2.2 were used for aspects of this research.
70
Figure 3.20 Optical image of 30 dpf polypropylene CCP™ fiber demonstrating shape retention
Figure 3.21 Scanning electron microscope image of bi-component fiber composed of EVOH sea and Nylon 6 island demonstrating shape retention
71
Figure 3.22 Scanning electron image of bi-component fiber composed of PLA sea and polypropylene islands which is not demonstrating correct shape integrity
3.2.2 Capillary Channel Polymer ™ Groove Analysis
Since particulates can have a diverse size distribution, it was necessary to optically examine
CCP™ fiber size, in particular, groove size to ensure that contaminants would physically be
able to fit within the grooves. Figures 3.23 – 3.27 show 3, 6, 15, 20, and 30 dpf SEM images
respectively. Since the most penetrating particle size (MPPS) during filtration is around .3
µm, these images show that these direct spun fibers have groove sizes which in theory could
accumulate a tremendous amount of particles. Figure 3.28 and 3.29 are images of bi-
component EVOH sea/Nylon 6 island CCP™ fibers, pre and post dissolution of the sea.
Once the sea is dissolved off, the island CCP ™ fibers are .22 dpf. The grooves of these bi-
72
component fibers measure approximately 2 µm, indicating the MPPS particle size would fit
into the grooves.
When building a depth filtration media it is important to be able to collect particles not only
on the surface of the filter but throughout the media. Depth filtration media provide for
high dirt holding capacity as well as longer service life due to the three dimensional structure.
Since one of the goals of this project is to increase collection efficiency of HEPA filters,
which includes collecting of particles varying in size, it was speculated that filers made with
CCP™ fibers should include fibers varying in size as well. As seen in figures 3.23 -3.29, these
fibers have varying groove sizes which may contribute towards collecting particles of
different sizes.
Figure 3.23 3 dpf CCP™ fiber with groove measuring approximately 8-9 μm
73
Figure 3.24 6 dpf CCP™ fiber with grooves measuring approximately 12 μm and 16-17 μm
Figure 3.25 15 dpf CCP ™ fibers with grooves measuring approximately 17 μm
74
Figure 3.26 20 dpf CCP ™ fiber grooves measuring approximately 17 µm and 12 μm
Figure 3.27 30 dpf CCP ™ fiber with groove measuring approximately 24 μm
75
Figure 3.28 Bi-component CCP™ fiber pre-dissolution of the sea, grooves measuring around 2 μm
Figure 3.29 Bi-component CCP™ fiber post-dissolution of the sea
76
3.2.3 Salt particle deposition study
Vaughn 26 suggested the benefits of CCP™ fibers for use in filtration due to particles
preference to collect within the grooves of the fibers, although, there were no analytical tests
conducted to support this hypothesis. To further investigate this theory and for the purposes
of this research, salt particles were sprayed onto nonwoven media composed of CCP™
fibers in order to examine particle deposition and loading trends on the fibers and within the
grooves. These samples were tested on a PALAS MFP Filter Test System at two different
flow rates (8.3 or 31.8 L/Min) and deposited with salt particles ranging 0.05-2 µm in size.
Figures 3.30 and 3.31 show the salt particles adhering within the grooves as well as on the
legs of the shaped fibers. These images indicate the particles not only fit on the sides of the
groove walls but also adhere deep within the grooves as seen in figure 3.31. These electron
microscope images show physical evidence that CCP™ particle collection sites are unique as
compared to round fibers and may assist in postponing the blinding off the filter media.
77
Figure 3.30 CCP ™ fiber loaded with salt particles at 2,000x magnification
Figure 3.31 CCP™ fiber loaded with salt particles at 5,000x magnification
78
3.2.4 Shape Factor
Shape factor, defined in equation 3, has been used to correlate shape and fabric properties59.
This equation characteristically relates the perimeter and area of the fiber cross-section in a
way where fibers with thin, long legs, with a larger perimeter, will have a high shape factor.
Perfectly round fibers will have a shape factor of 1. Table 3.2 lists shape factors for common
sized CCP™ fibers used during this research as well as a comparison to round 15 dpf fibers.
These measurements show that the CCP™ fibers have more than twice the shape factor of a
round fiber.
[equation 3]
Where
P = cross-sectional perimeter of the fiber (µm)
A = fiber cross-sectional area (µm)2
Table 3.2 Shape factor for round and CCP™ fibers
Fiber Shape Factor (η)
15 dpf Round 1.01
.22 dpf CCP™ bicomponent islands 2.80
3 dpf CCP™ 2.76
6 dpf CCP™ 2.52
15 dpf CCP™ 2.90
30 dpf CCP™ 2.80
79
3.3 Modeling of CCP™ Fibers
A modeling simulation study was conducted by Cox and Buckingham71 in conjunction with
analytical testing in order to simulate and design a HEPA filter with low pressure drop. The
model, which was based on the Stoke‘s equation, assumes the fluid is Newtonian since it is
being modeled (in air), assumed fluid incompressible, indicating the Reynold‘s number is
sufficiently small. Other assumptions used in this modeling were the particles are spherical,
fibers are parallel, cross-section of the media is uniform, no interaction of particles, no
electrostatic interaction between particle and media, as well as no particle buildup. To model
the fluid flow, the Langevin equation was used. Langevin describes the situation where
deterministic forces and stochastic forces are acting simultaneously on a particle in motion71.
In addition, the mechanisms of filtration, as discussed in section 1.8, are also assumed for
the modeling71. The Stoke‘s equation [equation 4] is defined as:
[equation 4]
Where:
= particle density
= particle radius
= Cunningham slip correction factor
= inflow velocity
= fluid viscosity R = average distance from the centroid of the fiber to the fiber boundary
80
3.3.1 Single fiber efficiency modeling
The single fiber efficiency modeling is defined as the measure of the number of particles
moving towards a fiber that the fiber will collect. The single fiber efficiency for a round fiber
is defined as:
[equation 5]
where
y = vertical distance from the center of the fiber to the point at which a particle will no longer be captured by the fiber R = radius of the fiber
Figure 3.32 Single fiber efficiency of round fiber71
Since the definition of single fiber efficiency for shaped fiber is not the same, single fiber
efficiency for shape had to be generalized as:
[equation 6]
81
where
Z = vertical distance from the point at which particles will flow above the fiber to the point at which particles will flow below the fiber R = average distance from the centroid of the fiber to the fiber boundary
Figure 3.33 Single fiber efficiency for CCP™ 71
Single fiber efficiency for CCP™ fibers was compared to round with varying density of
particles in kilograms per cubic meter, table 3.3, with a radius fixed of 0.2 µm. Since
efficiency of CCP™ fibers depends on their orientation, measurements over a variety of
orientations were considered and averaged. Results seen in figure 3.34 indicate CCP™ fibers
have a higher efficiency than that of a round. Figure 3.35 shows how the variation in CCP™
fiber orientation can affect efficiency with both CCP™ fibers having higher efficiency than
the round fiber. A second simulation was run with a fixed Stoke‘s number of 0.3, figure 3.36.
Once again every trial consistently showed the CCP™ fibers outperforming the round fibers
in terms of single fiber efficiency. For both runs, air velocity was set at 0.0748 meters per
82
second; air viscosity was 1.802 x 10-5 kilograms per meter per second and temperature at 293
kelvin.
Figure 3.34 Single fiber efficiency results of round versus CCP™ fiber with varying particle
density at a fixed radius of 0.2µm71
Table 3.3 Varying density of particle in kilograms per cubic meter of round versus CCP™ single fiber efficiency modeling.
Radius Density
(kilograms per m^3)
0.1 34806
0.2 11496
0.3 5673
0.4 3374
83
Figure 3.35 Single fiber efficiencies of round and CCP™ fibers in horizontal and vertical orientations71
84
Figure 3.36 Single fiber efficiency results of round versus CCP™ fiber with varying particle density with a fixed Stoke’s number of 0.371
3.3.2 Modeling to experimental validation of CCP™ fibers
In order to ensure to modeling results are valid with respect to experimental results, a
comparison between a simulation and experimental results were examined. Information
about the 3D filter was provided and needed to be converted to the 2D modeling
simulation. Cross sectional area was converted by the following equation [7],
[equation 7]
where
Acs is cross sectional area of the fiber d is denier of the fiber р is density of the material
85
The density of the fiber was obtained by dividing the basis weight of the filter by its
thickness. The packing density was determined by dividing the density of the filter by the
density of the polymer fiber.
SEM images of nonwoven filter media, figure 3.37, composed of CCP™ fibers of varying
denier plus round low melting temperature binding fibers were used to produce a 2D
modeling domain, figure 3.38. Since the simulation is in 2D it was not possible to reproduce
the exact 3 dimensional image. The filter in the simulation had the same packing density,
ratio of fiber types by number, and cross sectional area of each fiber type as the filter used in
the experiment. The simulation assumed a uniform or near-uniform distribution of fibers.
Figure 3.37 SEM cross section image of nonwoven filter media composed of CCP™ fibers, of
varying denier, and low melting PET binding fibers used for simulation study
86
Figure 3.38 Constructed simulation domain based off of SEM images of nonwoven filter
media composed of CCP™, of varying denier, and low melting PET binding fibers71
The inflow velocity was fixed at 0.0748 meters per second and particles sizes were 0.03 and
0.4 µm. The efficiencies obtained from the simulation were compared to the efficiencies
from several experiments. The results given in figure 3.39 and figure 3.40 show all
experimental results and the average of the experimental results respectively. The results
from the simulation fall within the range of experimental results, and that the results from
the simulation come reasonably close to the experimental average. Typically, HEPA filters
are thicker with a significantly higher number of fibers. These filters are generally too large
to model with the current design without making unreasonable assumptions, although, there
are ongoing steps to increase the number of the domains capable of being modeled.
87
Figure 3.39 Comparison of simulation to experimental data of collection efficiency of CCP™
filter media71
Figure 3.40 Comparison of simulation to experimental average of collection efficiency of
CCP™ filter media71
88
3.3.3 Pressure drop modeling of CCP™ fibers
A modeling study was conducted to compare the pressure drop of filter media composed of
CCP™ fibers to filter media composed of round fibers. The modeling domain was
constructed with the same number of fibers and identical cross-sectional area for each shape.
The input parameters for this simulation are listed in table 3.4. Since the pressure was set at
zero at a point on the inflow boundary, which is the left side, the value of the pressure on
the outflow boundary was negative. Figures 3.41 and 3.42 display the calculated pressure
drop of the CCP™ filter media and round filter media, respectively. These figures show a
model that theorizes a calculated reduced pressure drop of the filter media composed of
CCP™ fibers of having less than half the pressure drop as the filter media composed of
round fibers18.
Table 3.4 Input parameters for modeling pressure drop of filter media made of CCP™ fibers to round fibers
Face Velocity
Air viscosity
Number of particles
Particle Diameter
Particle Density
Time step
1.5 cm/s 1.75x10-5 Pa-s 30 0.15 µ 1000 kg/m3 2x10-7s
89
Figure 3.41 Simulated pressure drop of CCP™ fiber filter media in Pascals18
Figure 3.42 Simulated pressure drop of round fiber filter media in Pascals18
90
The modeling simulation was also able to show available paths that the particles took during
flow. Figures 3.43 and 3.44 shows the paths the particles took within the CCP™ filter media
and the round filter media, respectively. The paths highlighted in red are particles which
exited through the upper domain and then reentered on the lower side. Initial and terminal
points for each particle are denoted with an open circle to easily identify and are not to scale.
These figures show that particles are being collected throughout the domains and not on the
surface. This simulation is representative of a depth filtration system and not cake or surface
filtration.
Figure 3.43 Simulation of flow path of particles in a filtration media composed of CCP™ fibers18
91
Figure 3.44 Simulation of flow path of particles in a filtration media composed of round fibers18
3.4 Round Fiber versus CCP™ Fiber Filter Media
The rationale for including shaped fibers in filtration media was to offer the benefits of
increased dust collection capacity. However, although CCP™ fibers provide deep grooves
for dirt loading capability the CCP™ fibers manufactured during the scope of this research
tended to be much larger in size than traditional HEPA filter made with round fibers due to
limitations in technology and equipment. Due to this fact, these shaped fibers were not
expected to provide HEPA quality collection efficiency when used alone in a filter system.
In addition, the manufacturing methods available to make nonwovens filter media made
from CCP™ fibers involved a thermal bonding technique which reduced surface area at the
92
point bonding areas. These bonding areas will reduce flow rate making the practical side of
making filters somewhat different to the virtual filter approach used in the modeling.
3.4.1 Comparison of Pressure Drop of Round versus CCP ™ Filter Media
To compliment the simulation study on pressure drop, experimental data was collected
comparing CCP™ filter media to round filter media. Since it has been established that denier
for denier CCP fibers have a greater surface area than round fibers26, pressure drop
differences were looked at comparing 8 layered 15 dpf PP CCP™ nonwoven filter media to
8 layered 15 dpf round nonwoven filter media. Variations in basis weights within individual
sample layers must be noted due to the manufacturing methods. The results tended to be
scattered because of this variability but the overall trend is noted.
Four samples of CCP™ and 4 samples of the round fibers were measured on the Capillary
Flow Porometer (CFP) for pressure drop. The results can be seen in figure 3.45 and appear
to show no difference in pressure drop between these two samples. The average basis weight
of the 8 layer CCP™ filter media was 235 g/m2 (+/- 12.0%) and the 8 layer round filter
media measured 322 g/m2 (+/- 12.9%). Even though the same amount of layers was used,
the difference in basis weight occurred because of the ability of the round fibers to pack
closer than CCP™ fibers.
93
Figure 3.45 Comparison of flow rate (L/Min) versus differential pressure (PSI) of 8 layered 15 dpf CCP™ and round nonwoven filter media
The pressure drop of the data from figure 3.45 was examined again but this time taking into
consideration the fact the basis weights were not equal. Figure 3.46 shows the results of
pressure drop of the CCP™ versus round filter media when the data is normalized by
examining a ratio of flow rate to basis weight versus pressure drop. This graph demonstrates
the advantages of reduced pressure drop of the CCP™ filter media compared to the round
when they have equal weights.
0
20
40
60
80
100
120
140
0 0.5 1 1.5 2 2.5 3
Flo
w R
ate
(L
/M
in)
Pressure Drop (PSI)
Comparison of Pressure Drop of 8 layered 15 dpf CCP to Round Fibers
15 dpf CCP
15 dpf Round
94
Figure 3.46 Comparison of pressure drop of CCP™ versus round fiber filter media with normalized weights
Pressure drop differences were also studied comparing the 15 dpf CCP™ to 15 dpf round
fibers but creating equal surface areas by adjusting the average basis weight. The basis
weight was lowered in the CCP™ nonwovens by reducing the number of layers in the
nonwoven media. Table 3.5 lists the basis weights and surface area of the tested samples.
The approximate error for the CCP™ weights is 10% and 13% for the round. This test was
run twice, once where the CCP™ composite has 3 layers and again for a composite made
with 4 layers. It was not possible to produce CCP™ nonwoven material with the exactly the
same surface area as the round fiber composite by the layering technique. The 3 layered
CCP™ composite has 3.30 cm2 less surface area than the round and the 4 layers had 3.84
cm2 more surface area than the composite made with round fibers. Actual sample area tested
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0 0.5 1 1.5 2 2.5 3
No
rmali
zed
Flo
w R
ate
(L
*G
/M
in*M
2)
Pressure Drop (PSI)
Comparison of Pressure Drop of CCPtm vs Round Fibers with Normalized Weights
15 dpf CCP
15 dpf Round
95
was 1.2 cm in diameter. When determining surface area, 80% of the basis weight was used
since 20% of the weight was composed of binding fibers.
Figures 3.47 and 3.48 shows the pressure drop differences where the total surface areas of
the CCP™ and round filter media were measured to be comparable. The pressure drop of
the filter media made with round fibers was in general greater than the pressure drop of the
filter media made with CCP™ fibers. All data is based on 4 samples for CCP™ 3 layers,
CCP ™ 4 layers, and round. These results compliment the simulation model indicating the
filter media composed of CCP™ fibers have advantages in providing additional surface area
for particle collection while maintaining pressure differential compared to a filter media
composed of round fibers.
Table 3.5 Surface area and basis weight of 15 dpf round fiber filter media and 15 dpf CCP™ fiber filter media
CCP™ 3
layers CCP™ 4 layers Round
Average Basis Weight
(g/m2) 84 116 322
Calculated Average
Fiber Surface Area of
Tested Sample Area
(cm2) 32 39 35
96
Figure 3.47 Comparison of flow rate (L/Min) versus differential pressure (PSI) of filter media made from 3 layers of 15 dpf CCP™ and round fibers with comparable surface area
Figure 3.48 Comparison of flow rate (L/Min) versus differential pressure (PSI) of filter media made from 4 layers 15 dpf CCP™ and round fibers with comparable surface area
0
20
40
60
80
100
120
0 0.5 1 1.5 2 2.5 3
Flo
w R
ate
(L
/M
in)
Differential Pressure (PSI)
Comparison of Pressure Drop of CCPTM (3 layers) to Round Fibers with Comparable Surface Area
CCP Fibers
Round Fibers
0
20
40
60
80
100
120
0 0.5 1 1.5 2 2.5 3
Flo
w R
ate
(L
/M
in)
Differential Pressure (PSI)
Comparison of Pressure Drop of CCPTM (4 layers) and Round Fibers with Comparable Surface Area
CCP Fibers
Round Fibers
97
3.4.2 Experimental collection efficiency of filter media composed of CCP™ fibers
The collection efficiencies of the filter media containing CCP™ fibers were investigated
using the TSI fractional efficiency tester. The results obtained as shown in figure 3.49 were
for filter media composed of 15dpf PET CCP™ fibers, 80% by weight. This media also
contained 20% by weight of 4 dpf round PET binder fibers in order to aid bonding in the
calendaring process. The basis weight of the 6 layers sample was measured at 116 g/m2.
Figure 3.49 shows the flow rate (L/Min) versus collection efficiency of this media against
particles 0.3 µm in size. These results support the general trend that decreasing flow rate
allows for higher collection efficiency since the particles have more time within the media to
adhere to a fiber. Even at the lowest flow rate, 15.04 L/Min the collection efficiency of the
0.3 µm particles was around 80%. Thus this media does not come within accepted HEPA
standards. Perhaps this is not too surprising considering the size of the CCP™ fibers.
Figure 3.49 Collection efficiency of filter media composed of 80% 15 dpf CCP™ fibers
0
20
40
60
80
100
120
0 10 20 30 40 50 60 70 80 90
Flo
w R
ate
L/
Min
Collection Efficiency %
Collection Efficiency of 15 dpf CCPTM Filter Media
(0.3 µm particles)
98
The pressure drop, measured by mmH2O, was also analyzed as a function of increased flow
rate and results showed expected trend that as flow rate increases so does the resistance,
figure 3.50. This can be related to Darcy‘s Law [equation 8].
[equation 8]
where
= velocity
µ = fluid viscosity
= permeability
= applied pressure drop per unit thickness
Figure 3.50 Pressure drop of filter media made from 80% 15 dpf CCP ™ fibers
Collection efficiency was analyzed again for filter media made with CCP™ fibers but this
time with the fibers varying in denier. The composition of the media is detailed in table 3.6.
0
20
40
60
80
100
120
0 2 4 6 8 10 12
Flo
w R
ate
L/
Min
Resistance mmH20
Pressure Drop of 15 dpf CCPTM Filter Media
99
The round binding fibers were added, 20% by weight, to the total amount of all CCP™
fibers. Four, 4 layered samples as well as 1, 16 layer sample was tested for collection
efficiency against particles 0.03-0.40 µm in size at a flow rate of 15 L/Min. The basis weight
for a single layer measured approximately 28 g/m2. Figure 3.51 displays the results of
collection efficiency tests with none of the layers coming within the HEPA or ULPA
efficiency range. The 16 layer sample performed better than the 4 layer sample except for the
0.3 µm particle. More layers meant a higher amount of fibers as well as an increased tortuous
path for the particles to come into contact with. All samples were the same except for the
amount of layers. For the production runs produced during this research, there was
variability in basis weight within the single layer and multi layers within the calendared
nonwoven filter media. This variability caused inconsistencies and higher degree of air
permeability within certain areas from sample to sample. Higher permeability in turn allowed
for less particle collection.
Table 3.6 Filter composition of CCP™ filter media varying in denier
Denier per Filament (dpf) Shape Percentage (%)
6 CCP™ 55
15 CCP™ 40
30 CCP™ 4
4 Round (binding agent) 20
100
Figure 3.51 Collection efficiency of filter media composed of varying sized CCP™ fibers (graph generated by John Larzelere)
Figure 3.52 graphically represents the collection efficiency of filter media composed of 6, 15,
and 30 dpf CCP™ fibers as well as M98 filter media. As with all CCP™ filter media, 80% by
weight are of the fibers are CCP™ fibers and 20% by weight are binding fibers. All media
was tested at ~8.0 and ~32 L/Min flow rates. None of the media composed of CCP™
fibers performed comparable to the M98 media. The best performing CCP™ filter media
was the 6 dpf 8 layered sample at a flow rate of 8.0 L/Min which had approximately 70%
efficiency with particles in the size range of .05 - 0.4 µm.
0
10
20
30
40
50
60
0 0.1 0.2 0.3 0.4 0.5
Eff
icie
nc
y (
%)
Particle Size (um)
Collection Efficieny of Filter Media Composed of Varying Size CCPTM Fibers at Flow Rate of 15 L/Min
4 layer
4 layer
4 layer
4 layer
16 layers
101
Figure 3.52 Collection efficiency of CCP™ and M98 media at varying particle diameters and flow rates
The filter media composed of 15 dpf fibers performed better in terms of collection
efficiency as compared to the mixed denier filter media. There were manufacturing issues
with the 6 dpf fibers of the mixed denier nonwovens. During the carding process the 6 dpf
fibers experienced a high amount of fall out since the fibers were so small. This fall out may
have contributed to reduced web consistency which in turn impacted collection efficiency.
It was never envisaged that filter media composed of CCP™ fibers would alone work as a
HEPA filter since the fibers are too large. The results support this suggestion and therefore
no further collection efficiency data was looked at for filter media composed of a majority
CCP™ fibers alone.
0
20
40
60
80
100
120
0 0.2 0.4 0.6 0.8 1
Co
llecti
on
Eff
icie
ncy (%
)
Particle Diameter (µm)
Collection Efficiency of CCPTM and M98 Filter Media
15 dpf 4 Layers at 31.8 L/Min
15 dpf 4 Layers at 8.30 L/Min
30 dpf 4 layers at 31.8 L/Min
30 dpf 8 layers at 8.3 L/Min
6 dpf 4 layers at 31.7 L/Min
6 dpf 8 Layers at 8.3 L/Min
M98 at 8.3 L/Min
M98 at 33.4 L/Min
102
3.5 Meltbown Fibers
In addition to CCP™ fibers, this research also examined meltblown fibers for their
contribution towards improving HEPA and ULPA filtration. As previously discussed,
metlbown fibers offer a variety of advantages as a filter material as compared to electrospun
filter media; production rate and cost are two of the more appealing benefits. Advancements
in technology have also led to smaller meltblown fibers which alone do not provide the
structural integrity for a filter media but used in conjunction with other nonwovens or with a
substrate may provide additional surface area with low cost and manufacturing benefits. The
facility at Hills Inc. was used to fabricate several research level quantities of meltblown
nonwovens and the production variables can be found in Appendix A. This Appendix
describes the manufacturer‘s methods and polymers used. The characteristics of the fabrics
produced are listed in Table 3.7.
Table 3.7 Meltblown nonwovens manufactured by Hills Inc.
Sample ID Polymer Basis Weight
(g/m2) PE Scrim
Weight Average of Meltlbown
Assuming Scrim = ~ 8 (g/m2)
Standard Deviation
111808-01 Polypropylene 10.82 No 10.82 0.74
111808-02 Polypropylene 4.63 No 4.63 0.40
111808-03 Polypropylene 14.86 Yes 6.86 0.33
111808-04 Polypropylene 11.62 Yes 3.62 0.10
111808-05 Polypropylene 10.49 Yes 2.49 0.28
111808-06 Polypropylene 10.79 No 10.79 0.75
111808-07 Polypropylene 13.21 Yes 5.21 0.48
111808-08 Polypropylene 11.43 Yes 3.43 0.74
111808-09 Polypropylene 10.72 Yes 2.72 0.09
111808-10 Polypropylene 13.13 Yes 5.13 0.34
103
111808-11 Polypropylene 11.16 Yes 3.16 0.27
111808-12 Polypropylene 11.44 Yes 3.44 0.10
3.5.1 Meltblown fiber size analysis
Outside of the listed differences in Table 3.7 in basis weights and scrim, these meltblown
nonwovens were examined under microscopy imaging for fiber size distribution.
Advancements in meltblown technology have allowed for a modified meltblown fiber with
fiber sizes smaller than conventional fibers. Fiber size measurements were taken to attempt
to make correlation between the fiber size distribution, basis weights, and scrim to pressure
drop and efficiency. The following SEM images in this section of work are selected
representations of the nonwoven materials at a variety of magnifications. Measurements
were taken from all fibers visible within the individual SEM micrographs. Figure 3.53
graphically represents fiber size distribution for the 12 meltblown samples. Fiber sizes were
grouped into 4 categories; .01-.1 µm, .101-.5 µm, .501-1.0 µm, and 1.01 µm and higher. The
1.01 µm and higher category includes backing scrim which measured around ~17-22 µm.
104
Figure 3.53 Fiber size distribution graphs for meltblown fibers manufactured by Hills Inc.
SEM images were taken of the meltblown samples to look at the overall construction of the
material and take measurements of the fibers. Images were taken of samples 111808-01
through 111808-04 and 111808-06. Additionally samples 111808-01, 111808-02, and 111808-
06 were looked at since they were the only samples with no backing scrim. Samples 111808-
03 and 111808-04 were looked at since they contained a backing. Figures 3.54 – 3.57 are
images of sample 111808-01. This sample had a an average basis weight of 10.82 g/m2 which
was the highest of the meltblown fibers measured (not including backing). Typically,
meltblown fibers have diameters 1-4 mciron in size13 but as seen in figure 3.57 the fibers are
measuring in as small as 63 and 184 nm, respectively.
0.00%
10.00%
20.00%
30.00%
40.00%
50.00%
60.00%
70.00%
0.01 - 0.1 0.101 - 0.5 0.501 - 1 1.01 - above (includes backing
fibers)
Fiber Size (µm)
Fiber Size Distribution111808-01
111808-02
111808-03
111808-04
111808-05
111808-06
111808-07
111808-08
111808-09
111808-10
111808-11
111808-12
105
Figure 3.54 111808-01 at 2,000x magnification showing overall construction of nonwoven
Figure 3.55 111808-01 at 2,200x magnification showing detail of fiber size variability
106
Figure 3.56 111808-01 at 11,000x magnification detailing fiber size differences
Figure 3.57 111808-01 fibers measuring 63 and 184 nm
107
Samples 111808-02 images can be seen in figure 3.58 and 3.59. These images show the range
of fiber sizes within a very small sample. Figure 3.59 shows an individual fiber measuring at
180 nm.
Figure 3.58 111808-02 at 1,500 magnification showing variability in fiber diameters
108
Figure 3.59 111808-02 fiber measuring 180 nm
Figures 3.60 through 3.62 are images of 111808-3. This sample contains a polyethylene
backing which can be seen in figure 3.60 as the large fibers in the foreground of the image.
In comparison, figure 3.60 shows the reverse side of the nonwoven material with meltblown
fibers in the foreground and the backing fibers in the background. Figure 3.62 shows a fiber
for this sample measuring at 80 nm.
109
Figure 3.60 111808-03 with large backing fibers (~17-22 µm)
Figure 3.61 111808-03 at 5,000x magnification
110
Figure 3.62 111808-03 at 70,000x magnification with fiber measuring 80 nm
Figures 3.63 through 3.65 are of sample 111808-04. Figures 3.63 and 3.65 show opposite
sides of the nonwoven material where the backing can be seen in figure 3.63 clearly and in
the background of the meltblown fibers in figure 3.64. Figure 3.65 image is another example
of how varying fiber sizes can be in the nonwoven material.
111
Figure 3.63 111808-04 at 450x magnification with backing fibers visible in foreground
Figure 3.64 111808-04 at 500x magnification with backing fibers in background
112
Figure 3.65 111808-04 at 4,000 magnification
Figures 3.66 and 3.67 are of sample 111808-06. No backing is used for this sample. Figure
3.67 shows a fiber measuring in at 161 nm. Once again, these images show the extent of
fiber size variability as well as the representation of nano-size fibers in some of the media.
All samples, excluding 111808-04 through 111808-06, showed some nano-size fibers less
than 100 nm in diameter in their construction. Although no fibers less than 100 nm were
seen in any of the SEM images of 111808-04 through 111808-06, there remains the
possibility of nano-fibers within the media that did not show up through the imaging
conducted during this research.
113
Figure 3.66 111808-06 at 2,500x magnification with varying fiber sizes
Figure 3.67 111808-06 with fiber measuring at 161 nm
114
3.5.2 Pressure drop and collection efficiency of meltblown fibers
After viewing fiber size distribution characteristics of the meltblown samples, efficiency and
pressure drop data was collected for the samples using the TSI fractional efficiency tester.
The graph in figure 3.68 shows collection efficiency (%) versus flow rate (L/Min) for single
layer samples. It was found that single layer samples have higher permeability and lower
collection efficiency than layered samples.
Samples 111808-06 and 111808-01 had the highest collection efficiency, respectively. They
were two of the three meltblown samples with no scrim. The third sample with no scrim,
111808-02, had a much lower basis weight, 4.63 g/m2 as compared to 10.79 and 10.82 g/m2
of the other two non-scrim samples. This may have contributed to the reduced efficiency
since the fiber size distribution of this media did not demonstrate a majority of fibers in the
nano-size range indicating there may have been fewer fibers which resulted in less surface
area for particle collection.
115
Figure 3.68 Collection efficiency versus flow rate for 111808-xx single layer meltblown samples
Figure 3.69 graphically shows the percentage of fibers that measured less than 0.5 µm in size.
Media composed with a higher percentage of these fibers could potentially demonstrate
higher collection efficiency and/or a decreased pressure drop since they would have more
fibers performing within the slip flow regime. Samples 111808-02, 111808-07, and 111808-
06 had the three highest amounts of fibers within the slip flow regime, respectively. When
comparing this to their collection efficiency, it is noted that 111808-02 had the lowest basis
weight (4.63 g/m2) of all 12 samples but it also contained no scrim. No scrim allows for filter
media with a higher percentage of fibers which are sub-micron in size which may provide for
increased surface area for particle collection. Sample 111808-07 had the second highest
percentage of fibers measuring less than 0.5 µm in size and also had the third best collection
0
10
20
30
40
50
60
70
8.26 8.27 8.28 8.29 8.3 8.31 8.32 8.33 8.34
Co
llecti
on
Eff
icie
ncy %
Flow Rate L/Min
Collection Efficiency of 111808-xx Meltblown Single Layer Samples
111808-01
111808-02
111808-03
111808-04
111808-05
111808-06
111808-07
111808-08
111808-09
111808-10
111808-11
116
efficiency. This sample did have a scrim backing which may have attributed to slightly
decreased collection efficiency since a percentage of the fibers were larger and may not have
contributed the increased surface area that a non-scrim filter media did. Sample 111808-06
had the third highest amount of fibers within the slip flow regime, 10.79 g/m2 of meltblown
and without scrim. This combination of no scrim and a high percentage of meltblown fibers,
especially less than 0.5 µm in size, showed the highest collection efficiency. Excluding
111808-09, it appears that samples having greater than 4 g/m2 of meltblown have better
efficiency than the samples with meltblowns containing less than 4 g/m2. It is probable
sample 111808-09 experienced damage since the individual layers are so fragile resulting in
extremely low collection efficiency. Samples 111808-01 and 111808-06 had the two highest
amounts of meltblown fibers, 10.82 g/m2 and 10.79 g/m2 respectively, both with no scrim.
These samples also had the top two best collection efficiencies. A higher percentage of
meltblown provides smaller fibers which allows for more surface area.
117
Figure 3.69 Percentage of meltblown fibers measured less than 0.5 µm in size
Figure 3.70 shows the collection efficiency versus resistance of the same single meltblown
samples. The graph represents a general trend that increased collection efficiency comes with
increased pressure drop. Sample 111808-06 was the exception to this general rule. This also
may be attributed to its unique combination of no scrim and a high percentage of fibers
within the slip flow regime. Some samples registered a negative resistance by the instrument
as a result of its limitations in measuring pressure drop so low. The pressure drops for single
layer samples were exceptionally low and the negative numbers are relative for each sample.
0.00%
10.00%
20.00%
30.00%
40.00%
50.00%
60.00%
70.00%
80.00%
Percentage of Metlblown Fibers Less Than 0.5µm In Size111808-01
111808-02
111808-03
111808-04
111808-05
111808-06
111808-07
111808-08
111808-09
111808-10
111808-11
111808-12
118
Figure 3.70 Collection efficiency versus pressure drop for 111808-xx single layer meltblown samples
Based on the initial efficiency data collected, sample 111808-06 was layered and tested again
since it showed the highest efficiency with single layer. The graph in figure 3.71 shows the
results after sample 111808-06 was layered and collection efficiency was recorded. The
sample was used to prepare two, four, and eight layered composite samples which were
tested at approximately 8.3 and 32.0 L/Min except for the eight layer sample which was
tested at 16.6 L/min instead of 32.0L/Min. The general trend indicates more layers of
meltblown samples results in higher collection efficiency. The eight layer sample with a flow
rate of 8.331 L/Min had the highest efficiency of 99.97%. The eight layer sample tested at a
flow rate of 16.61 L/Min had the second efficiency of 99.85%. It should be noted that
standard testing condition for M98 media is at 8.3 L/Min. Slower flow rates tend to give
0
10
20
30
40
50
60
70
-0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8 1 1.2
Co
llecti
on
Eff
icie
ncy %
Resistance mmH2O
Collection Efficiency vs Pressure Drop of 111808-xx Meltblown Single Layer Samples 111808-01
111808-02
111808-03
111808-04
111808-05
111808-06
111808-07
111808-08
111808-09
111808-10
111808-11
111808-12
119
higher efficiency ratings since the particles have a longer exposure time within the media as
compared to a higher flow rate where particles have less time to adhere to a fiber by one of
the filtration mechanisms.
Figure 3.71 Collection efficiency of 111808-06 multi layered meltblown samples
Layered 111808-06 samples run at 8.3 L/Min were examined for collection efficiency versus
resistance. The trend of increased pressure differential with increased collection efficiency is
demonstrated in figure 3.72. The pressure drop per layer is reasonably linear.
50
55
60
65
70
75
80
85
90
95
100
0 5 10 15 20 25 30 35
Co
llecti
on
Eff
icie
ncy %
Flow Rate L/Min
Collection Efficiency of 111808-06 Layered Samples
1 layer
1 layer
2 layer
2 layer
4 layer
4 layer
8 layer
8 layer
120
Figure 3.72 Collection efficiency versus pressure drop for 111808-06 layered samples at 8.3 L/Min flow rate
3.5.2.1 Comparison of collection efficiency and pressure drop of M98 media to 111808-06
media
Since metlbown sample, 111808-06, demonstrated collection efficiency within HEPA
standards and a goal of this research was to improve on the M98 standards, the two samples
were compared for collection efficiency and pressure drop. Figure 3.73 is a graph
demonstrating collection efficiency for both samples at a flow rate of 8.3 L/Min. The graph
shows the general trend of decreased collection efficiency with most penetrating size
particles. Table 3.8 provides the details of these results with the meltblown sample having
slightly higher collection efficiency at 99.965% compared to 99.961% for the M98 filter
media. The meltblown sample did register a slightly higher resistance of 8.067 mmH2O
0
20
40
60
80
100
120
0 1 2 3 4 5 6 7 8 9
Co
llecti
on
Eff
icie
ncy %
Resistance mmH2O
Collection Efficiency vs Pressure Drop for 111808-06 Layered Samples at 8.3 L/Min Flow Rate
1 layer
2 layer
4 layer
8 layer
121
compared to the M98 media‘s resistance of 6.378 mmH2O. The metlbown sample was
layered up 6 time making the basis weight comparable to the M98 filter media. When
comparing the percentage of fibers for these two samples within the slip flow range, 111808-
06 had a higher amount at 61% compared to 43% for M98 media. Figure 3.65 graphically
compares fiber size distribution in detail for the two samples. Based on the amount of fibers
within the slip flow regime, the meltblown sample was expected to have a lower pressure
drop than the M98 material. Overall, the meltblown did have slightly higher pressure drop
but better efficiency, in particular, with particles below 0.2 µm in diameter.
Figure 3.73 Collection efficiency of 111808-06 meltblown sample versus M98 filter media at 8.3 L/Min flow rate
Table 3.8 Collection efficiency and resistance results of meltblown sample versus M98 filter media
Sample Basis Weight
(g/m2) MPPS (µm) Collection
Efficiency (%) Resistance (mmH2O)
111808-06 66.96 0.2 99.965 8.067
M98 69.69 0.126 99.961 6.378
99.955
99.960
99.965
99.970
99.975
99.980
99.985
99.990
99.995
100.000
0.00 0.20 0.40 0.60 0.80 1.00
Co
llecti
on
Eff
icie
ncy %
Particle Diameter (µm)
Collection Efficiency of Meltblown vs M98 Filter Media
111808-06
M98
122
3.5.2.2 Comparison of physical properties of M98 media to 111808-06 media
M98 filter media has a paper type quality to it, besides for being light in weight it also is
considerably thin. Since the meltlbown sample 111808-06 was performing comparable or
slightly better in collection efficiency to the M98 filter media based on basis weight, the
thickness of the sample was analyzed to see how it compared. Figure 3.74 is an SEM image
of single layer of the 111808-06 meltblown sample where the thickness measured 362 µm.
Figure 3.75 is an SEM image of the M98 filter media where the thickness measured 600 µm.
In order for the meltblown layers to have comparable collection efficiency, they were layered
6 times which would make their thickness much higher than the M98 media which was
tested as a single layer. Maximizing space may be an important parameter when installing
filter systems since certain filters can be located in tight quarters such as submarines. HEPA
filters are commonly pleated to obtain the utmost surface area. If sample thickness is too
high then issues with pleating the media could arise.
123
Figure 3.74 Thickness of single layer of meltblown sample 111808-06 measuring 362 µm
Figure 3.75 Thickness of M98 filter media measuring 600 µm
124
A comparison of fiber size distribution of 111808-06 to M98 was analyzed. Figure 3.76
shows a representation of the respective fiber sizes for the two filter media. The one clear
difference between the two samples is that the meltblown media measurements indicated no
fibers diameters in the .01-.10 range. A majority of the fibers for both filters were in the
.101-.500 range.
Figure 3.76 Comparison of fiber size distribution of meltblown to M98 filter media
3.6 Bi-component Fibers
Bi-component fibers were examined for their ability to be processed into submicron
diameter fibers comparable in size to glass fibers but also offer the flexibility of being unique
in shape and polymer. Several types of bi-component fibers were examined and are listed in
table 3.9. All fibers listed in this table were designed to be post-processed via dissolvable or
hydrolysis methods that would be appropriately environmentally acceptable to the textile
industry. As such, the preferred sheath polymers used were either water or alkali soluble.
0%
10%
20%
30%
40%
50%
60%
70%
.01-0.1 .101-0.5 0.501-1.00 1.01-above
Fiber size in Microns
Fiber Size Distribution of Meltblown vs M98 Filter Media
111808-06
M98
125
The rational for using this separation process was so the filter media could be made by a
wet-laid system which is a common method used in manufacturing of filter media. Ideally
there would be no need for additional equipment or major changes in formation technique
and thus if possible this route would be an attractive method for media production. The
separation process expected for these bi-component fibers was dissolution/hydrolysis.
Details of the manufacturer and dissolution parameters are discussed in further detail in the
forthcoming sections.
Table 3.9 List of bi-component fibers analyzed for use in filter media
Bicomponent Cross Section Polymer (%) Separation Process
Capillary Channel Polymer Islands Polypropylene 19 Dissolvable
Round Sea Polylactic Acid 81
Capillary Channel Polymer Islands Nylon 6 19 Dissolvable
Round Sea EXCEVAL (PVA) 81
Capillary Channel Polymer Islands Nylon 6 19 Dissolvable
Round Sea ESPET 81
Snowflake Islands PET 40 Dissolvable
Round Sea WSPET 60
Snowflake Islands Nylon 6 25 Dissolvable
Round Sea EXCEVAL (PVA) 75
Snowflake Islands Nylon 6 20 Dissolvable
Round Sea ESPET 80
126
156K Islands Nylon 6 10 Dissolvable
Round Sea ESPET 90
3.6.1 Bi-component Fiber Characterization of Soluble PET Fibers
Bi-component fibers composed of soluble PET sea were evaluated for fiber characterization
via FTIR, TGA and DSC analytical techniques. ESPET (Easily Soluble PET) is
manufactured by Kuraray Inc. and is said to undergo alkali hydrolysis68. WSPET is
manufactured by Eastone and is defined as a water soluble bi-component fiber. Fiber
characterization began with the ESPET, easy soluble PET sea, and WSPET, water soluble
PET sea, for composition and thermal analysis.
3.6.2.1 Fourier Transform Infrared (FTIR) Spectroscopy
FTIR was used to determine whether the WSPET sea or ESPET sea reflected characteristic
peaks of PET fibers. In addition, this method was used to determine if there were any
significant differences in spectrum between these two fiber types. Two different fibers
examined were, 40/60 PET/WSPET and 20/80 Nylon 6/ESPET. Since the FTIR
instrument is for surface analysis and the sea encompasses the islands, it was assumed the
spectrum produced would be of the sea and not of the islands.
Figure 3.77 is a spectra of the 40/60 PET/WSPET fiber and figure 3.78 of the 156,000
islands-in-the-sea composed of 20/80 Nylon 6/ESPET. Not only do both spectrums show
the characteristic peaks of PET fiber but they indicate little or no difference in composition
between the two of them. In addition, these spectra were compared to the library database
which confirmed the closest match to PET.
127
Figure 3.77 IR spectra of bi-component 40/60 PET/WSPET snowflake fiber
128
Figure 3.78 IR spectra of bi-component 156,000 islands-in-a-sea fiber composed of 80/20 ESPET sea/Nylon 6 island
3.6.1.2 Thermogravimetric analysis (TGA)
The ESPET and WSPET sea fibers were also subjected to TGA analysis. Since hot washing
or thermal bonding may be needed to complete the filter media manufacturing process,
thermal degradation was examined. Figures 3.79 and 3.80 are TGA spectrums of 40/60
PET/WSPET snowflake fibers and 10/90 Nylon 6/ESPET 156,000 islands-in-the-sea
fibers. Degradation for both fibers begins around 400°C. This temperature is much higher
than any thermal bonding (highest temperature used for calendaring was 130°C) or attempts
to dissolve out the sea, which were used for the nonwoven manufacturing during the scope
129
of this research project.
Figure 3.79 TGA analysis of 40/60 WSPET/PET snowflake fibers heated at 20°C per minute
130
Figure 3.80 TGA analysis of 10/90 Nylon 6/ESPET 156,000 islands-in-a-sea fiber heated at 20°C per minute
3.6.1.3Differntial Scanning Calorimetry (DSC)
Differential scanning calorimetry (DSC) was used to analyze crystallanity, glass transition,
(Tg) and melting temperature(Tm), of the bi-component fibers. Figures 3.81 through 3.84
show the thermograms of the 40/60 PET/WSPET snowflake and 10/90 Nylon 6/ESPET
156,000 islands-in-the-sea fibers before and after quench, respectively. Since the snowflake
bi-component fiber contained both PET polymers phases, phase transitions were expected
to be similar for the thermograms. Figure 3.80 shows the fiber, before quench, with four
endothermic peaks, one large at 57.6°C due to physical aging, a secondary peak at 70.4°C,
131
and a third even smaller peak around 80°C. This thermogram shows that these fibers have a
very complicated thermal history. These Tg and physical aging peak indicate a glass
transition of the as spun fiber to be lower than the Tg of 80°C for PET69 homopolymers.
The fourth peak at 248.9°C is more indicative of the melting temperature of PET which has
a listed range of 260-265°C69. Once again, this peak is slightly lower than PET
homopolymers but maybe due to copolymerization. Figure 3.81 is a thermogram of the fiber
rerun after quench and shows a crystallization peak. It shows a crystallization peaks at
134.8°C since the sample was quench cooled in an amorphous state allowing for re-
crystallization to occur upon heating above the Tg.
Figure 3.81 DSC thermogram of 40/60 PET/WSPET snowflake fiber before quench
132
Figure 3.82 DSC thermogram of 40/60 PET/WSPET snowflake fiber after quench
Figures 3.83 and 3.84 are DSC thermograms of the 10/90 Nylon 6/ESPET islands-in-the-
sea. Figure 3.83 shows an endothermic peak due to physical aging of the ESPET at 64.3°C.
There are two endothermic peaks indicating melting temperature, a smaller peak at 211.0°
for Nylon 6 and a larger secondary peak at 230.4°C for ESPET sea. Once again this is
slightly lower than listed PET melting temperatures but it likely due to a modified molecular
backbone that aids in alkali hydrolysis for ESPET. Figure 3.83 indicates crystallization peak
at 135.0°C which occurs on reheating after the quench cool procedure.
133
It should be noted that the sea components for both sets of fibers were at a much higher
percentage than the islands (particularly with the 156,000 islands-in-the-sea fibers) and so the
sea polymers dominate the thermal responses of the materials in controlled heating regimes.
Figure 3.83 DSC thermogram of 10/90 Nylon 6/ESPET 156,000 islands-in-the-sea fiber before quench
134
Figure 3.84 DSC thermogram of 10/90 Nylon 6/ESPET 156,000 islands-in-the-sea fiber after quench
3.6.2 Bi-component SEM Imaging and Dissolution
The bi-component fibers were examined with SEM imaging in order to get a sense of
geometry and fiber size. Dissolution of the sea was also investigated since the sea
components of the fibers needed to be dissolved off in a timely and relatively easy manner as
to expose the submicron island fibers. Preferentially, the solvent should be water or non-
toxic and non-corrosive so as not to cause any damage to wet-laid equipment. As shown in
figure 3.85, the CCP™ PP island/PLA sea were not considered in this research since the
shape integrity of the CCP™ fibers was incorrect.
135
Figure 3.85 CCP™ PP island/PLA sea bi-component fiber without shape integrity
3.6.2.1 Nylon 6 CCP™ Islands/Poly(vinyl alcohol) (PVA) sea Bi-component Fibers
Exceval™ is a water soluble polymer created by Kuraray America, Inc. and is sold as an
ethylene modified copolymer of polyvinyl alcohol. Nylon 6 CCP™ island/PVA copolymer
sea bi-components were examined under SEM pre dissolution of the sea to verify shape
retention of CCP™ fiber. Figure 3.86 shows a cross section of a fiber. The CCP™ islands
are clearly defined and the shape integrity is intact. Poly (vinyl alcohol) is a water soluble
polymer so dissolution of the sea was expected to occur with heated water. Figure 3.87 is an
image of the islands post dissolution which occurred with heated tap water at 100°C for 20
minutes and lowered to 70°C for an additional 20 minutes. These fibers offered great
potential for use in filter manufacturing since the shape is well defined and the size, 0.22 dpf,
is an order of magnitude smaller than shaped fibers produced via direct melt spinning
136
techniques. Unfortunately, these fibers could not be thoroughly investigated as a viable
option since the quantity necessary to pass through the non-woven equipment at Clemson is
larger than the production runs made for these fibers. The reason for this is that the
Exceval™ polymer begins to degrade during bi-component melt extrusion after
approximately an hour due to the residence time in the complex bi-component melt
spinning pack.
Figure 3.86 Nylon 6 CCP™ island/Poly(vinyl alcohol) sea fiber cross section at 3,200x magnification pre dissolution of the sea
137
Figure 3.87 Nylon 6 CCP™ fibers post sea dissolution at 2,700x magnification
3.6.2.2 Fibers containing easily soluble or hydrolysable polyester sea materials
PET soluble sea fibers were also examined. The sea, ESPET, is described as an easy soluble
polyester fiber and WSPET, as water soluble polyester. As with the Exceval™ sea, these
polymers would be used to make filters via wet-laid process making it a viable industry
manufacturing option. These fibers were imaged to analyze fiber diameter and dissolution
parameters were investigated for proper solvent concentration.
3.6.2.2.1 Snowflake PET Islands/WSPET Sea with Heated Water as Solvent
Snowflake fibers made of 40% PET islands and 60% WSPET sea were investigated for use
in a wet-laid process. Since the island was described as a water soluble PET, they were
initially subjected to heated DI water in order to attempt to dissolve out the sea. SEM
138
images were taken of side on view of the snowflake fibers. Figure 3.88 is an image of the
snowflake fibers prior to any dissolution in hot water. Figure 3.89 is an image of the fibers
after being subjected to 60°-70° C water and agitated with a stir bar for approximately 2
hours. These fibers no longer appear to be as smooth as the fibers prior to heated water
treatment but the sea is not dissolved off. Figure 3.90 is an image after the fibers were
immersed in 80° - 90° C water and agitated with a stir bar for approximately 3 ½ hours. It
became obvious that this heated water procedure was inadequate in terms of completely
dissolving away the WSPET sea.
Figure 3.88 Snowflake WSPET sea fibers prior to any attempt to dissolve out the sea
139
Figure 3.89 Snowflake WSPET sea fibers after 2 hours in 60° - 70°C water treatment
Figure 3.90 Snowflake WSPET sea fibers after being immersed in 80° -90°C water for approximately 3.5 hours
140
3.6.2.2.2 1Snowflake PET Islands/WSPET Sea with Non-Water Solvents and Sonication
Since heated water alone was not able to dissolve off the water soluble sea, different solvents
were investigated. Table 3.10 lists the solvents which were used to attempt to dissolve off
the sea completely. The Triton x-114, a surfactant, was used to try and cause further
separation of the fibers within the nonwoven media since they fibers appeared to be
agglomerating. The fibers were subjected to the solvent in a heated sonicator for 2.5 hours.
The temperature was kept around 65°C in order to keep it below the Tg of the PET islands.
Figures 3.91-3.93 are SEM images of these fibers after being subjected to the aqueous
mixture of 1500 ml DI water 0.11% Triton x-114 and 4% sodium sulphate at 65°C.
Although there is evidence the sea is beginning to dissolve off it is not completely eliminated
nor are the island fibers isolated as they continue to show a great amount of agglomeration
even with the surfactant present. Ideally, the island fibers would be separated and the filter
media would have a much more open structure. Figure 3.93 shows that nano-fibers are
visible in part of the media with an individual fiber measuring 210 nm. Not all fibers will
uniformly measure 210 nm; there will be variations within the fiber diameter. Based on post
drawn measured yarn denier of 223, island fibers should have an approximate dpf of .005.
Table 3.10 Dissolution solvents used to attempt to dissolve off WSPET sea with PET islands
Snowflake Island
Snowflake Sea Solvent
40% PET 60% WSPET
1700 ml DI water .11% Triton x-114
2.33% Sodium hydroxide @ ~65°C
141
40% PET 60% WSPET
1500 ml DI water .11% Triton x-114
.4% Sodium sulphate @ ~65°C
Figure 3.91 Snowflake WSPET sea fibers at 300x magnification after exposure to solvents and sonicator for 2.5 hours
142
Figure 3.92 Snowflake WSPET sea fibers at 470x magnification after exposure to solvents and sonicator for 2.5 hours
Figure 3.93 Snowflake PET island fiber measuring 210 nm after exposure to solvents and sonicator for 2.5 hours
143
Since dissolution of the bi-component fibers composed of WSPET sea and PET islands was
unsuccessful in any reasonable time frame the focus was shifted to the fibers composed of
the ESPET sea but with Nylon 6 islands.
3.6.2.2.3 Snowflake Nylon 6 islands/ESPET Sea
Bi-component fibers made of 20% Nylon 6 islands and 80% ESPET sea were subjected to
the parameters listed in table 3.11 in order to attempt to dissolve off the sea. These bi-
component fibers were subjected to NaOH, an alkali solvent, which should dissolve off the
ESPET sea by attacking the ester groups of the PET without damaging the Nylon islands.
Fibers were placed in a 250 ml jar along with the respective solvents and sonicated, at 60°,
for varying cycles. The fibers were rinsed under tap water, dried, and viewed with SEM
imaging to verify if dissolution of sea was successful. The butyl alcohol was used in
conjunction with the NaOH since it acts as a plasticizing/wetting agent. Figures 3.94 and
3.95 are images of the snowflake fibers with the highest percentage of NaOH as well as with
and without butyl alcohol, respectively. Even though once again, there is evidence of the sea
being dissolved off there appears to be some sea residual left on the island fibers. Figure 3.96
shows that islands could measure around 200 nm.
Table 3.11 Dissolution solvents used to attempt to dissolve off ESPET sea with Nylon 6 islands
Snowflake Island
Snowflake Sea
Weight (grams)
Solvent Cycles
(30 minutes in sonicator)
20% Nylon 6 80% ESPET 0.1 4% NaOH 1 (cold start)
20% Nylon 6 80% ESPET 0.1 4% NaOH 1
20% Nylon 6 80% ESPET 0.2 4% NaOH 1
144
20% Nylon 6 80% ESPET 0.2 27% NaOH 1 (cold start)
20% Nylon 6 80% ESPET 0.2 27% NaOH 1
20% Nylon 6 80% ESPET 0.2 27% NaOH 2
20% Nylon 6 80% ESPET 0.2 27% NaOH 3
20% Nylon 6 80% ESPET 0.2
4% NaOH 4% Butyl alcohol 1
20% Nylon 6 80% ESPET 0.2
27% NaOH 4% Butyl alcohol 2
20% Nylon 6 80% ESPET 0.2
27% NaOH 8% Butyl alcohol 3
Figure 3.94 Snowflake 20/80 Nylon 6/ESPET fibers after 3 cycles in 27% NaOH
145
Figure 3.95 Snowflake 20/80 Nylon 6/ESPET fibers after 3 cycles in 27% NaOH and 8% butyl alcohol
Figure 3.96 Snowflake Nylon 6 islands measuring 229 nm
146
Although the NaOH solvent used in conjunction with the wetting agent was helping remove
the ESPET sea, the effort and time necessary to make the process work were somewhat
discouraging.
3.6.2.2.4 156,000 Islands-in-the-sea bi-component fibers
Bi-component 156,000 islands-in-the-sea fibers composed of 10% Nylon 6 islands and 90%
ESPET sea were investigated next for their potential use in wet-laid filter. These high island
count fibers where on dissolution expected to produce fibers in the sub 100 nm range and
therefore could offer other possibilities to compete with electrospinning processes. Figures
3.97 – 3.100 are SEM images of the 156,000 islands-in-the-sea fibers pre dissolution of sea.
The images in figures 3.97 – 3.98 expose the nano-sized islands within the sea. Figure 3.100
shows an individual nano-fiber island measuring 44 nm. The hope was if the fibers were
allowed to spend more time in the dissolution mixture then the NaOH would hydrolyze the
PET sea to leave only the Nylon nano-fiber islands. The potential then would be to make
either a complete filter composed entirely of nano-sized fibers or to have such fibers
dispersed within a wet laid system.
147
Figure 3.97 156,000 islands-in-a-sea bi-component fiber at 1,500x magnification where sea is not dissolved off but islands are exposed
Figure 3.98 156,000 islands-in-a-sea bi-component fiber at 1,800x magnification where sea is not dissolved off but islands are emerging
148
Figure 3.99 156,000 islands-in-a-sea bi-component fiber at 4,000x magnification where sea is not dissolved off but islands are visible
Figure 3.100 Single ‘island’ in 156,000 islands-in-a-sea bi-component fibers measuring 44nm
149
Table 3.12 list times and solvent concentrations used to attempt to completely remove the
ESPET sea on the 156k islands-in-the-sea fibers. It was thought since the percentage of sea,
90%, was so high, that an aggressive combination of solvent, heat, and time would be
required to make a filter media composed only of the nano-size islands. Samples were placed
in a 250 ml jar with de-ionized water and, unless otherwise noted, days in sonicator were for
8 hour periods as the procedure was shut down over night (non continuous). Treatments
were also performed under a continuous regime whereby the fibers were subjected to
solvents, heat and sonciation (Table 3.12).
Table 3.12 Dissolution parameters used to attempt to dissolve off sea of 156K islands-in-the-sea fiber
156K Islands Round Sea Weight (grams)
Solvent
Time in Sonicator at
60° C (days)
10% Nylon 6 90% ESPET 1.0
27% NaOH 4% Butyl alcohol 1
10% Nylon 6 90% ESPET 1.0
27% NaOH 8% Butyl alcohol 5
10% Nylon 6 90% ESPET 1.0
27% NaOH 8% Butyl alcohol 7
10% Nylon 6 90% ESPET 1.0
27% NaOH 8% Butyl alcohol 3*
10% Nylon 6 90% ESPET 1.0
27% NaOH 8% Butyl alcohol 5*
150
10% Nylon 6 90% ESPET 1.0
27% NaOH 8% Butyl alcohol 7*
* continuous 24 hour sonication
Figures 3.101 – 3.107 are SEM images showing the progression of dissolution attempts by
increasing the time which the fibers were exposed to solvents, heat, and sonication. Once
again, it appeared that sea was dissolving off but the nano-fibers islands never became
completely separated from the sea. There always remained a layer of polymer sea on the
nano-sized fibers. Figures 3.106 and 3.107 show that even after the fibers were placed in the
sonicator for 7 days under a continuous setting, remains of the ESPET sea are evident and
the island fibers start to experience breakage. If these fibers were left any longer than 7 days
in the sonicator there was no physical evidence of the fibers remaining, only a layer of cloudy
water on the bottom the jars which were unable to be filtered out.
151
Figure 3.101 156K islands-in-the-sea fibers after 1 day in sonicator and solvents
Figure 3.102 156K islands-in-the-sea after 5 days in sonicator (non-continuous) and solvents
152
Figure 3.103 156K islands-in-the-sea after 7 days in sonicator (non-continuous) and solvents
Figure 3.104 156 islands-in-the-sea after 3 days in sonicator (continuous) and solvents
153
Figure 3.105 156 islands-in-the-sea after 5 days in sonicator (continuous) and solvents
Figure 3.106 156 islands-in-the-sea after 7 days in sonicator (continuous) and solvents at 1,000x magnification
154
Figure 3.107 156 islands-in-the-sea after 7 days in sonicator (continuous) and solvents at 11,000x magnification
Due to the resistant nature of the ESPET sea with the current solvent system, an additional
effort of dissolving off the sea was made with the addition of benzyl alcohol in place of the
butyl alcohol. Benzyl alcohol will act a plasticizer lowering the Tg allowing the NaOH to
better penetrate the polymer chains of the sea. Table 3.13 lists the parameters used for
dissolution attempts.
Table 3.13 Dissolution parameters used to attempt to dissolve off sea of 156K islands-in-the-sea fiber using benzyl alcohol
156K Islands Round Sea Weight (grams)
Solvent
Time in Sonicator at
60° C (hours)
10% Nylon 6 90% ESPET 1.0 27% NaOH
1% Benzyl alcohol 32
155
10% Nylon 6 90% ESPET 1.0 27% NaOH
1% Benzyl alcohol 56
10% Nylon 6 90% ESPET 1.0 27% NaOH
1% Benzyl alcohol 68
10% Nylon 6 90% ESPET 1.0 27% NaOH
1% Benzyl alcohol 92
Figures 3.108 through 3.111 show the progressive attempts to dissolve off the ESPET sea.
Needle-like fibers begin to emerge from the sea but as previously experienced at no point
does the sea become completely dissolved off.
Figure 3.108 156 islands-in-the-sea after 32 hours in solvent with the addition of benzyl alcohol
156
Figure 3.109 156k islands-in-the-sea after 56 hours in solvent with the addition of benzyl alcohol
Figure 3.10 156k islands-in-the-sea after 68 hours in solvent with the addition of benzyl alcohol
157
Figure3.111 156K islands-in-the-sea after 92 hours in solvent with the addition of benzyl alcohol
Since one of the goals of this research was to investigate alternative methods of making
submicron fibers for use in filtration but maintaining industrial production rates, removal of
the sea component in a timely and easy method was vital. Unfortunately, all studies
conducted in the scope of this dissertation proved unsuccessful for these terms.
3.7 Composite Filter Media
Composite filter media was investigated for its ability to provide layers of fibers varying in
size and shape which could offer advantages in filtration. Since CCP™ fibers offer
advantages in terms of additional surface area, and meltblown nonwovens offer benefits in
terms of speed of fabric production as well as submicron size fibers, the theory of a
composite filter made of meltblown and nonwovens made with CCP™ fibers was analyzed.
158
Figure 3.112 is a schematic of this composite filter media with meltblown layers sandwiched
between layers of CCP™ fibers varying in denier. The investigator theorized as the
contaminants are carried by air flow through the filtration media, larger sized particulates
would enter the first layer of the filter media and be collected by the CCP™ fibers. The
smaller particulates would be collected by the submicron sized fibers within the meltblown
layers. This depth filtration system could potentially postpone the expiration of the filter all
the while providing HEPA quality collection efficiency. As well the design is symmetrical
which would allow for easier installation. The exact composition of the two filter samples
made and tested are described in table 3.14. The major difference between the two samples
is that composite 1 contained nylon CCP™ fibers which were 3, 6, 15, and 30 dpf and
composite 2 contained polypropylene CCP™ fibers which were all 15 dpf.
Air Flow
CCP Mixed dpf Layers
Meltblown Layers
CCP Mixed dpf Layers
Figure 3.112 Schematic of composite filter media composed of CCP™ and metlblown layers
159
Table 3.14 Composition of composite filter media made with CCP™ fibers and meltblown nonwovens
Composite 1 Composite 2
Layers 1-4
Nylon 6 CCPTM 3 dpf (6.2%)
CCPTM 6 dpf (31.3%) CCPTM 15 dpf (25.0%) CCPTM30 dpf (37.3%)
Polypropylene CCPTM 15 dpf
Layer 5
Polypropylene Meltblown 7 gsm
Polypropylene Meltblown 7 gsm
Layers 6 -9
Nylon 6 CCPTM 3 dpf (6.2%)
CCPTM 6 dpf (31.3%) CCPTM 15 dpf (25.0%) CCPTM30 dpf (37.3%)
Polypropylene CCPTM 15 dpf
Figure 3.113 shows the collection efficiency of composite filter media at three different flow
rates, 8.3, 15.1, and 31.8 L/Min. None of the samples demonstrated collection efficiency
within HEPA or ULPA standards. When the two samples are tested at the same flow rates,
their collection efficiencies are almost identical. As expected, both samples demonstrated the
highest efficiency, around 75% when tested at the slowest flow rate, ~8 L/Min. As with all
filter media, slower flow rates allow for particles to have more time to come in contact with
a fiber and be collected. The composite media‘s collection efficiency was slightly higher
(~75%) as compared to the 8 layered 6 dpf CCP™ media of ~70% but lower than
meltblown sample 111808-06 filter media which was 99.965%
160
Figure 3.114 shows the pressure drop of the two composite filter media. Even though both
filters had extremely low pressure drop, Composite 1, which contains CCP™ fibers varying
in denier and smaller in size than composite 2, exhibits the higher pressure drop,
approximately 2.5 mmH2O. Composite 2 contains all 15 dpf CCP™ fibers which may
provide a more open filter structure allowing for a lower pressure drop of approximately 1.9
mmH2O.
In addition to the composite filters being unable to collect particulates at an acceptable
efficiency, the thickness as compared to M98 media is much larger. Further work will be
needed on composite design to improve collection efficiency and lower pressure drop. The
composites examined during this research only used a single layer of meltblown sandwiched
between multiple layers of CCP™ nonwovens. Future composites could include additional
layers of meltblown as well as potentially using shaped fiber spun bond which was not
available at the time of this research.
161
Figure 3.113 Collection efficiency of composite filter media with varying flow rates
Figure 3.114 Pressure drop of composite filter media
0
10
20
30
40
50
60
70
80
0 0.1 0.2 0.3 0.4 0.5
Co
llecti
on
Eff
ien
cy (
%)
Particle Diameter (µm)
Collection Efficiency of Composite Filter Media
Composite 1 8.32 Flow Rate
Composite 2 15.1 Flow Rate
Composite 1 31.8 Flow Rate
Composite 2 8.3 Flow Rate
Composite 2 15.1 Flow Rate
Composite 2 31.8 Flow Rate
0
5
10
15
20
25
30
35
0 0.5 1 1.5 2 2.5
Flo
w R
ate
(L
/M
in)
Resistance (mmH2O)
Pressure Drop of Composite Filter Media
Composite 1
Compsite 2
162
3.7.1 Figure of Merit (FOM) of Composite Media
Filter quality has been related to figure of merit (FOM) which is defined as follows:
[equation 9]
Where:
P = Penetration (1- efficiency)
∆p = Pressure drop
Although FOM measurements are not the only method of determining quality filter, it is a
metric used by certain industries, including the funding organization of this research. Others
factors, such as filter thickness may play a part, particularly in pleated filters. FOM
measurements of composite media composed of meltblown and CCP™ fibers were analyzed
for purposes of this research against M98 filter media. These results were based off of initial
small R&D samples produced by Hills Inc. Based on these results, further exploration into
meltblown layers for use in filtration were investigated and those explorations were discussed
in Section 3.5. Tables 3.15 lists the polymer, shape and basis weights of the individual layers
which make up the composite filter media. Table 3.16 provides a description of the
composition of layers with their cumulative basis weights. The results of the flow rate versus
FOM are shown in figure 3.104.
Of the samples tested Clemson MB7 demonstrated the best results with a FOM value of
14.38 as compared to the M98 media value of 14.07. The major difference between Clemson
MB7 from the other composite materials was that it was a completely meltblown media
containing no layers made with CCP™ fibers and also included a scrim backing. None of the
163
other samples contained a scrim backing. The basis weight being 98 g/m2 was higher than
the M98‘s basis weight of 70 g/m2.
FOM calculations do not take into consideration the balance of efficiency and pressure drop.
Samples could demonstrate exceptionally low pressure drop and collection efficiency and
score a high FOM value. For this reason, FOM measurements were used as supplemental
data.
Table 3.15 Description of polymer, shape, and basis weights of FOM filter media
Sample ID Polymer Shape Basis Weight Minus
Scrim (g/m2)
PE Scrim = ~8 g/m2
Total Basis Weight (g/m2)
Clemson MB7 Polypropylene Round 6.00 Yes 14.00
99 031008-01 Polypropylene Round 7.00 No 7.00
Mixed 6/15/30 dpf PET CCP™ 28.5 No 28.5
Table 3.16 Composition of FOM filter media
Clemson M12 Clemson 3 M6B Clemson 3 TM6B Clemson MB7
Top Layer(s) None None 2 layers
Mixed 6/15/30 dpf None
Middle Layer(s) 12 layers
99 031008-01 6 layers
99 031008-01 6 layers
99 031008-01 7 Layers
Bottom Layer(s) None 2 layers
Mixed 6/15/30 dpf
2 layers Mixed 6/15/30 dpf
None
Basis Weight (g/m2) 84 99 156 98
164
Figure 3.115 Figure of merit measurements of meltblown samples versus M98 filter media (graph generated by John Larzelere)
Clemson 3 TM6B
Clemson 3 M6BClemson M12
M98
Clemson MB7
8.29
8.295
8.3
8.305
8.31
8.315
8.32
8.325
8.33
8.335
8.34
12 12.5 13 13.5 14 14.5
Flo
w R
ate
(L
/min
)
FOM (pascal)
Flow at 8.3 L/min vs FOM
Clemson 3 TM6B
Clemson 3 M6B
Clemson M12
M98
Clemson MB7
165
CHAPTER 4
CONCLUSIONS
This research examined unique geometry CCP™ fibers as well as submicron fibers
manufactured through metlblown and bi-component spinning processes for use in HEPA
and ULPA filtration. The purpose of this research was to investigate alternative methods to
electrospinning to produce submicron and nano-size fibers with their increased surface area
to potentially increase with collection efficiency and pressure drop. As well, this dissertation
delved into the theorized benefits of slip flow effect to ease pressure differential and
potentially increase collection efficiency with fibers measuring less than 500 nm in diameter.
4.1 Capillary Channel Polymers (CCP™) Fibers for Use in HEPA and ULPA
Filtration
Experimental and modeling data performed on filter media composed of CCP™ fibers
indicated their increased surface area as compared to round fibers of the same denier
provided additional particle collection sites. In particular, the deep grooves of the fibers can
collect particles varying in size dependent on the groove size which in turn is dictated by the
fiber‘s denier. A benefit of having particles preferentially adhere within the deep grooves is it
allows more time for the filter to function before becoming loaded. Extending filter life is
beneficial in terms of cost as well as minimizing labor time and effort required for changing
out filters. Another key benefit of using CCP™ fibers is they provide a tortuous path for the
particulates limiting face loading. These complex paths allow the CCP™ fibers to produce a
depth filtration tool which follows in the tradition of HEPA and ULPA filter media.
166
Particulates collect within the media and not only on the surface which helps in postponing
the expiration of the media.
Due to limitations in manufacturing CCP™ fibers smaller in size than those spun through
traditional melt spun processes; collection efficiency of CCP™ media was subpar when
compared to standard HEPA filter media. CCP™ fibers tested during the term of this
research project were large in size compared to the M98 media fibers. Therefore, it was
believed they were unable to pack as closely allowing a high percentage of particulates to
flow through the media without being collected.
In terms of pressure drop, CCP™ fibers show advantages when compared to round fibers
of comparable denier. Filter media made from CCP™ had a lower pressure drop than a filter
media composed of the same denier round fiber. Reducing pressure drop is vital to
maximizing the life cycle of filters. Used alone in filter media, CCP™ fibers did not possess
the attributes necessary to improve on the current standards of HEPA and ULPA filtration.
4.2 Meltblown Nonwovens for Use in HEPA and ULPA Filtration
Meltblown nonwovens possess the ease of process and financial benefits desired for filter
media production. Since modified meltblown lines are capable of producing submicron and
nano-sized fibers, they showed great potential for filter media. When analyzed during this
research project they were scrutinized and compared against the current M98 media
standards. The meltbown filter media showed notable collection efficiency and pressure
drop data.
167
Fiber size distribution for meltblown and M98 media was analyzed for percentage of fibers
which fall within the slip flow regime (<.50 µm diameter). The meltblown sample which
contained a high amount of fibers within the slip flow regime and contained no scrim
demonstrated HEPA quality collection efficiency when compared to the M98 media with
comparable basis weight. This meltblown also had a slightly higher pressure drop.
Modified meltblown nonwovens show potential in providing HEPA quality collection
efficiency with reasonable pressure drop when the majority of fibers are measuring less than
0.50 µm. The media does not have the structural integrity to be used alone and would be
required to have additional support. As well, modifications to the media are necessary to
manipulate the thickness if expectations for M98 thickness are to stay as are. Since pleating
HEPA is standard, the paper like quality of the media is necessary and is not currently
feasible with the meltblown nonwovens examined during this research project.
4.3 Bi-component Fibers for Use in HEPA and ULPA Filtration
An appealing quality of bi-component fibers is their ability to be melt spun into unique
shape fibers as well as have a dissolvable separation process which results in submicron and
nano-size fibers. The ease and efficiency of the separation process is necessary in order for
bi-component fibers to compete with current filtration manufacturing processes.
Bi-component sheath/core fibers show the potential to compete with fibers produced by
indolent electrospun methods in terms of size. They have the advantage though to be
produced at a much faster speed as well as having unique geometries. One necessity of
manufacturing filter media is ease of process so they can be made via wet laid manufacturing
168
conditions. SEM images verified submicron and nano-sized island fibers produced from
WSPET and ESPET seas. Microscopy imaging also confirmed CCP™ islands,
approximately. 22 dpf in size are feasible through a bi-component method which would not
be possible through conventional melt spinning techniques.
Challenges in using bi-component fibers arose when attempts were made to process the sea
off in an efficient manner. No combination of heat, time, or solvent was discerned which
made it possible to completely dissolve off the sea. All attempts to expose the submicron
and nano-sized island fibers were negated by a layer of polymer sea remaining on the fibers.
Although bi-component fibers show considerable potential for future use as an alternative
method to electrospinning methods, considerable efforts into producing a consistent and
legitimately dissolvable sea must be made.
4.4 Composite Filter Media
A composite filter media made from CCP™ filter media and meltblown media in theory
provided hopes of creating the ideal HEPA or ULPA filter. Meltblown media with its
promising collection efficiency and CCP™ fiber media with their deep grooves and ample
surface area were anticipated to be a perfect combination. Unfortunately even though
pressure drop was low as compared to M98 media, the collection efficiency suffered. It was
theorized the denier of CCP™ fibers used during this research were large with respect to
M98 fiber size. These large shaped fibers provided an open tortuous path for air flow which
potentially aided in reducing pressure drop but also reduced the chances of contaminants
being collected by a mechanism of filtration. The composites tested in this research
contained only single layers of meltblown nonwovens and future attempts should consider
169
increasing the amount of meltblown fibers. In addition, even if these composites were at
HEPA level efficiency, the thickness would be much higher than M98 media making it
difficult to pleat and produce an acceptable and installable system.
This research only scratched the surface of the potential for using CCP™ fibers alone or in
a composite material for filtration. Due to limitation in manufacturing capabilities of bi-
component CCP™ fibers it is unclear if making these CCP™ fibers smaller could decrease
the tortuous air path enough to increase collection but maintain a lower pressure drop. All
evidence in this research supports the theory that CCP™ fibers could increase filter life by
providing additional real estate for particulates to adhere to. At minimum, filter media
composed of CCP™ fibers have the characteristics of viable pre-filter. These pre-filters
would remove larger contaminant from air flow prior to reaching a HEPA or ULPA filter,
extending their life cycle.
170
CHAPTER 5
FUTURE WORK
5.1 Unique Geometry Fibers
CCP™ shaped fibers were the primary focus of this research because of their increased
surface area (as compared to round of equal denier), unique geometry which includes deep
grooved channels, as well as ease and accessibility to manufacturing through melt spinning.
Although this shape proved an excellent candidate for investigation of unique geometry
fibers for use in filtration, it is not the only shape. For instance, the Cornell University
modeling study suggested that Y shaped fibers with relatively large area to volume ratio will
help with particle collection by Brownian motion54. Through the encouraging results
demonstrated with shaped fibers with their increased surface area thus far, it may prove
fruitful to investigate other unique geometry fibers. Figure 5.1 is an example of a bi-
component polypropylene core and PLA sheath fiber which may also prove successful in
demonstrating benefits in filtration. This shape has numerous grooves which may allow for
collection sites of particulates. With the advancements in melt spinning engineering and
spinnerets design, it is possible to create numerous uniquely shaped fibers. The possibilities
for use in filtration are endless to explore.
171
Figure 5.1 PP core/PLA sheath bi-component fiber for potential investigation for use in filtration media
5.1.1 Unique geometry fibers through meltblown process
At the time of this research, there was no available method of producing submicron shaped
fibers via a spunbond process. With recent advancements in spunbond process lines, it may
be possible to investigate a filtration manufacturing system where advantages of producing a
direct spun filter media made with CCP™ and other shaped fibers is possible. The
combination of fast production rate, a direct spun process, submicron and shaped fibers may
offer substantial benefits for producing a filter media.
5.2 Electrostatic charge
As discussed in section 1.7, electrostatic attraction is one of the four mechanisms of
filtration but was not addressed during the scope of this dissertation. One of the suggested
172
advantages of imparting charge to filtration media is the ability for the fibers to collect
particles at a further distance, with a more open structure. Unlike glass fibers, polymer fibers
are suitable for electrostatic charging. Triboelectric effect and corona treatment are two
methods of charging polymer fibers16 and as discussed in section 1.1.1 have demonstrated
advantages in filtrations. The advantages of using a filter media composed of electrostatically
charged CCP™ fibers is yet to be explored. The combination of CCP™ fibers with their
deep grooves where particles can collect in addition to a charge to facilitate the capture of
particles may be a substantial advantage in the filtration industry. In addition, since
nonwovens can be calendared into varying layers, it is possible to make alternating layers of
positively and negatively charged fibers. Figure 5.2 shows a schematic of this proposed filter
system.
Figure 5.2 Schematic of electrostatically charged nonwoven filter media
173
5.3 Modeling Simulation
Due to limitations in modeling simulation discussed in section 3.3, improvements are
necessary to properly represent a complete filtration system. Currently, the modeling
simulation is a two dimensional representation which does accurately model the three
dimensional characteristics of an actual filter. In addition to creating a three dimensional
filter; developing a more accurate simulation, such as thicker filters composed of smaller
fibers, can be improved by increasing storage. Other issues that need to be examined in
further detail include electrostatic charge on fibers or the system, eddy or vortex currents
within the grooves of the CCP™ fibers, as well as particle build up on the filters.
174
Appendix A
SAMPLE 111808-01
GENERAL CONDITIONS
Cross Section: HOMO NANO
Pack Build: 1960 100 HPI
Polymer B: 1800 PP Drying Time: n/a
Gm Hole Min: 0.0046
Run Speed: 1.75 mpm
Meter Pump Size: 2.92 cc/rev
Melt Pump Speed 4.2 RPM
Aspirator Pres: 4
Forming Table Distance: 8 in
Room Temp: F
Air heat set @ 320C, actual 259C.
No Scrim
175
EXTRUSION TEMPERATURES SAMPLE 111808-01
DEVICE Measured Value oC Setpoint oC 1. EXTRUDER A ZONE 1 TMP 181.0 180.0 2. EXTRUDER A ZONE 2 TMP 180.0 180.0 3. EXTRUDER A ZONE 3 TMP 180.0 180.0 4. EXTRUDER A ZONE 4 TMP 180.0 180.0 5. EXTRUDER A FLANGE TMP 200.0 200.0 6. EXTRUDER B ZONE 1 TMP 181.0 180.0 7. EXTRUDER B ZONE 2 TMP 180.0 180.0 8. EXTRUDER B ZONE 3 TMP 179.0 180.0 9. EXTRUDER B ZONE 4 TMP 180.0 180.0 10. EXTRUDER B FLANGE TMP 200.0 200.0 11. Inlet manifold flange TMP 262.0 - 12. MELT PUMP B HEATER - - 13. TRANSFER LINE A ZONE 1 TMP 200.0 200.0 14. TRANSFER LINE A ZONE 2 TMP 200.0 200.0 15. TRANSFER LINE A ZONE 3 TMP 200.0 200.0 16. TRANSFER LINE A ZONE 4 TMP 200.0 200.0 17. TRANSFER LINE B ZONE 1 TMP 230.0 230.0 18. TRANSFER LINE B ZONE 2 TMP 230.0 230.0 19. TRANSFER LINE B ZONE 3 TMP 230.0 230.0 20. TRANSFER LINE B ZONE 4 TMP 230.0 230.0 21. REGEN HEATER 19.0 0.0 22. SPIN BEAM ZONE 1 TMP 231.0 230.0 23. SPIN BEAM ZONE 2 TMP 230.0 230.0 24. SPIN BEAM ZONE 3 TMP 230.0 230.0 25. SPIN BEAM ZONE 4 TMP 230.0 230.0 26. SPIN BEAM ZONE 5 TMP 230.0 230.0 27. SPIN BEAM ZONE 6 TMP 230.0 230.0 28. SPIN BEAM ZONE 7 TMP 231.0 230.0 29. SPIN BEAM ZONE 8 TMP 230.0 230.0 30. SPIN BEAM AIR TMP - - 31. A TRANSFER LINE OUTLET TMP 185.0 - 32. B TRANSFER LINE OUTLET TMP 199.0 - 33. A MELT PUMP OUTLET TMP 233.0 - 34. B MELT PUMP OUTLET TMP 226.0 - 35. QUENCH AIR 1 TMP 20.0 - 36. INTERNAL PACK TEMP 233.0 - 37. AIR MANIFOLD IN TEMP 274.0 - 38. QUENCH AIR CONTROL TMP 22.0 5.0 39. PACK PREHEATER CONTROL TMP 20.0 -
176
SAMPLE 111808-02
GENERAL CONDITIONS
Cross Section: HOMO NANO
Pack Build: 1960 100 HPI
Polymer B: 1800 PP
Gm Hole Min: 0.0046
Run Speed: 3.5 mpm
Meter Pump Size: 2.92 cc/rev
Melt Pump Speed 4.8 RPM
Aspirator Pres: 4
Forming Table Distance: 8 in
Room Temp:
Air heat set @ 320C, actual 263C.
No Scrim
177
EXTRUSION TEMPERATURES SAMPLE 111808-02
DEVICE Measured Value oC Setpoint oC
1. EXTRUDER A ZONE 1 TMP 181.0 180.0 2. EXTRUDER A ZONE 2 TMP 180.0 180.0 3. EXTRUDER A ZONE 3 TMP 180.0 180.0 4. EXTRUDER A ZONE 4 TMP 180.0 180.0 5. EXTRUDER A FLANGE TMP 200.0 200.0 6. EXTRUDER B ZONE 1 TMP 180.0 180.0 7. EXTRUDER B ZONE 2 TMP 180.0 180.0 8. EXTRUDER B ZONE 3 TMP 180.0 180.0 9. EXTRUDER B ZONE 4 TMP 180.0 180.0 10. EXTRUDER B FLANGE TMP 200.0 200.0 11. Inlet manifold flange TMP 265.0 - 12. MELT PUMP B HEATER - - 13. TRANSFER LINE A ZONE 1 TMP 200.0 200.0 14. TRANSFER LINE A ZONE 2 TMP 200.0 200.0 15. TRANSFER LINE A ZONE 3 TMP 200.0 200.0 16. TRANSFER LINE A ZONE 4 TMP 200.0 200.0 17. TRANSFER LINE B ZONE 1 TMP 230.0 230.0 18. TRANSFER LINE B ZONE 2 TMP 230.0 230.0 19. TRANSFER LINE B ZONE 3 TMP 230.0 230.0 20. TRANSFER LINE B ZONE 4 TMP 230.0 230.0 21. REGEN HEATER 19.0 - 22. SPIN BEAM ZONE 1 TMP 231.0 230.0 23. SPIN BEAM ZONE 2 TMP 230.0 230.0 24. SPIN BEAM ZONE 3 TMP 230.0 230.0 25. SPIN BEAM ZONE 4 TMP 230.0 230.0 26. SPIN BEAM ZONE 5 TMP 230.0 230.0 27. SPIN BEAM ZONE 6 TMP 230.0 230.0 28. SPIN BEAM ZONE 7 TMP 231.0 230.0 29. SPIN BEAM ZONE 8 TMP 230.0 230.0 30. SPIN BEAM AIR TMP - - 31. A TRANSFER LINE OUTLET TMP 185.0 - 32. B TRANSFER LINE OUTLET TMP 201.0 - 33. A MELT PUMP OUTLET TMP 234.0 - 34. B MELT PUMP OUTLET TMP 228.0 - 35. QUENCH AIR 1 TMP 21.0 - 36. INTERNAL PACK TEMP 236.0 - 37. AIR MANIFOLD IN TEMP 276.0 - 38. QUENCH AIR CONTROL TMP 22.0 5.0 39. PACK PREHEATER CONTROL TMP 20.0 -
178
SAMPLE 111808-03
GENERAL CONDITIONS
Cross Section: HOMO NANO
Pack Build: 1960 100 HPI
Polymer B: 1800 PP Drying Time: n/a
Gm Hole Min: 0.0057
Run Speed: 3.5 mpm
Meter Pump Size: 2.92 cc/rev
Melt Pump Speed 5.2 RPM
Aspirator Pres: 5
Forming Table Distance: 8 in
Room Temp:
Air heat set @ 320C, actual 263C.
Polyethylene Scrim
179
EXTRUSION TEMPERATURES SAMPLE 111808-03
DEVICE Measured Value oC Setpoint oC
1. EXTRUDER A ZONE 1 TMP 180.0 180.0 2. EXTRUDER A ZONE 2 TMP 180.0 180.0 3. EXTRUDER A ZONE 3 TMP 180.0 180.0 4. EXTRUDER A ZONE 4 TMP 180.0 180.0 5. EXTRUDER A FLANGE TMP 200.0 200.0 6. EXTRUDER B ZONE 1 TMP 180.0 180.0 7. EXTRUDER B ZONE 2 TMP 180.0 180.0 8. EXTRUDER B ZONE 3 TMP 180.0 180.0 9. EXTRUDER B ZONE 4 TMP 180.0 180.0 10. EXTRUDER B FLANGE TMP 200.0 200.0 11. Inlet manifold flange TMP 271.0 - 12. MELT PUMP B HEATER - - 13. TRANSFER LINE A ZONE 1 TMP 200.0 200.0 14. TRANSFER LINE A ZONE 2 TMP 200.0 200.0 15. TRANSFER LINE A ZONE 3 TMP 199.0 200.0 16. TRANSFER LINE A ZONE 4 TMP 200.0 200.0 17. TRANSFER LINE B ZONE 1 TMP 230.0 230.0 18. TRANSFER LINE B ZONE 2 TMP 230.0 230.0 19. TRANSFER LINE B ZONE 3 TMP 230.0 230.0 20. TRANSFER LINE B ZONE 4 TMP 240.0 240.0 21. REGEN HEATER 19.0 - 22. SPIN BEAM ZONE 1 TMP 246.0 245.0 23. SPIN BEAM ZONE 2 TMP 246.0 245.0 24. SPIN BEAM ZONE 3 TMP 245.0 245.0 25. SPIN BEAM ZONE 4 TMP 245.0 245.0 26. SPIN BEAM ZONE 5 TMP 245.0 245.0 27. SPIN BEAM ZONE 6 TMP 245.0 245.0 28. SPIN BEAM ZONE 7 TMP 245.0 245.0 29. SPIN BEAM ZONE 8 TMP 245.0 245.0 30. SPIN BEAM AIR TMP - - 31. A TRANSFER LINE OUTLET TMP 187.0 - 32. B TRANSFER LINE OUTLET TMP 210.0 - 33. A MELT PUMP OUTLET TMP 24.0 - 34. B MELT PUMP OUTLET TMP 233.0 - 35. QUENCH AIR 1 TMP 22.0 - 36. INTERNAL PACK TEMP 245.0 - 37. AIR MANIFOLD IN TEMP 280.0 - 38. QUENCH AIR CONTROL TMP 23.0 5.0 39. PACK PREHEATER CONTROL TMP21.0 -
180
SAMPLE 111808-04
GENERAL CONDITIONS
Cross Section: HOMO NANO
Pack Build: 1960 100 HPI
Polymer B: 1800 PP Drying Time: n/a
Gm Hole Min: 0.0057
Run Speed: 7 mpm
Meter Pump Size: 2.92 cc/rev
Melt Pump Speed 5.2 RPM
Aspirator Pres: 4.5
Forming Table Distance: 8 in
Room Temp:
Air heat set @ 315C, actual 244C.
Polyethylene Scrim
181
EXTRUSION TEMPERATURES SAMPLE 111808-04
DEVICE Measured Value oC Setpoint oC
1. EXTRUDER A ZONE 1 TMP 181.0 180.0 2. EXTRUDER A ZONE 2 TMP 181.0 180.0 3. EXTRUDER A ZONE 3 TMP 180.0 180.0 4. EXTRUDER A ZONE 4 TMP 180.0 180.0 5. EXTRUDER A FLANGE TMP 200.0 200.0 6. EXTRUDER B ZONE 1 TMP 180.0 180.0 7. EXTRUDER B ZONE 2 TMP 179.0 180.0 8. EXTRUDER B ZONE 3 TMP 180.0 180.0 9. EXTRUDER B ZONE 4 TMP 180.0 180.0 10. EXTRUDER B FLANGE TMP 200.0 200.0 11. Inlet manifold flange TMP 267.0 - 12. MELT PUMP B HEATER - - 13. TRANSFER LINE A ZONE 1 TMP 200.0 200.0 14. TRANSFER LINE A ZONE 2 TMP 200.0 200.0 15. TRANSFER LINE A ZONE 3 TMP 200.0 200.0 16. TRANSFER LINE A ZONE 4 TMP 200.0 200.0 17. TRANSFER LINE B ZONE 1 TMP 230.0 230.0 18. TRANSFER LINE B ZONE 2 TMP 230.0 230.0 19. TRANSFER LINE B ZONE 3 TMP 230.0 230.0 20. TRANSFER LINE B ZONE 4 TMP 240.0 240.0 21. REGEN HEATER 19.0 - 22. SPIN BEAM ZONE 1 TMP 246.0 245.0 23. SPIN BEAM ZONE 2 TMP 245.0 245.0 24. SPIN BEAM ZONE 3 TMP 245.0 245.0 25. SPIN BEAM ZONE 4 TMP 245.0 245.0 26. SPIN BEAM ZONE 5 TMP 245.0 245.0 27. SPIN BEAM ZONE 6 TMP 245.0 245.0 28. SPIN BEAM ZONE 7 TMP 246.0 245.0 29. SPIN BEAM ZONE 8 TMP 245.0 245.0 30. SPIN BEAM AIR TMP - - 31. A TRANSFER LINE OUTLET TMP 187.0 - 32. B TRANSFER LINE OUTLET TMP 210.0 - 33. A MELT PUMP OUTLET TMP 243.0 - 34. B MELT PUMP OUTLET TMP 235.0 - 35. QUENCH AIR 1 TMP 21.0 - 36. INTERNAL PACK TEMP 244.0 - 37. AIR MANIFOLD IN TEMP 277.0 - 38. QUENCH AIR CONTROL TMP 23.0 5.0 39. PACK PREHEATER CONTROL TMP21.0 -
182
SAMPLE 111808-05
GENERAL CONDITIONS
Cross Section: HOMO NANO
Pack Build: 1960 100 HPI
Polymer B: 1800 PP Drying Time: n/a
Gm Hole Min: 0.0057
Run Speed: 14 mpm
Meter Pump Size: 2.92 cc/rev
Melt Pump Speed 14.2 RPM
Aspirator Pres: 6
Forming Table Distance: 8 in
Room Temp:
Air heat set @ 315C, actual 244C.
Polyethylene Scrim
183
EXTRUSION TEMPERATURES SAMPLE 111808-05
DEVICE Measured Value oC Setpoint oC
1. EXTRUDER A ZONE 1 TMP 181.0 180.0 2. EXTRUDER A ZONE 2 TMP 180.0 180.0 3. EXTRUDER A ZONE 3 TMP 180.0 180.0 4. EXTRUDER A ZONE 4 TMP 180.0 180.0 5. EXTRUDER A FLANGE TMP 200.0 200.0 6. EXTRUDER B ZONE 1 TMP 180.0 180.0 7. EXTRUDER B ZONE 2 TMP 181.0 180.0 8. EXTRUDER B ZONE 3 TMP 181.0 180.0 9. EXTRUDER B ZONE 4 TMP 180.0 180.0 10. EXTRUDER B FLANGE TMP 200.0 200.0 11. Inlet manifold flange TMP 271.0 - 12. MELT PUMP B HEATER - - 13. TRANSFER LINE A ZONE 1 TMP 200.0 200.0 14. TRANSFER LINE A ZONE 2 TMP 200.0 200.0 15. TRANSFER LINE A ZONE 3 TMP 201.0 200.0 16. TRANSFER LINE A ZONE 4 TMP 200.0 200.0 17. TRANSFER LINE B ZONE 1 TMP 230.0 230.0 18. TRANSFER LINE B ZONE 2 TMP 230.0 230.0 19. TRANSFER LINE B ZONE 3 TMP 230.0 230.0 20. TRANSFER LINE B ZONE 4 TMP 240.0 240.0 21. REGEN HEATER 19.0 - 22. SPIN BEAM ZONE 1 TMP 246.0 245.0 23. SPIN BEAM ZONE 2 TMP 246.0 245.0 24. SPIN BEAM ZONE 3 TMP 245.0 245.0 25. SPIN BEAM ZONE 4 TMP 244.0 245.0 26. SPIN BEAM ZONE 5 TMP 245.0 245.0 27. SPIN BEAM ZONE 6 TMP 245.0 245.0 28. SPIN BEAM ZONE 7 TMP 245.0 245.0 29. SPIN BEAM ZONE 8 TMP 245.0 245.0 30. SPIN BEAM AIR TMP - - 31. A TRANSFER LINE OUTLET TMP 187.0 - 32. B TRANSFER LINE OUTLET TMP 211.0 - 33. A MELT PUMP OUTLET TMP 24.0 - 34. B MELT PUMP OUTLET TMP 236.0 - 35. QUENCH AIR 1 TMP 22.0 - 36. INTERNAL PAK TEMP 244.0 - 37. AIR MANIFOLD IN TEMP 280.0 - 38. QUENCH AIR CONTROL TMP 24.0 5.0 39. PAK PREHEATER CONTROL TMP 22.0 -
184
SAMPLE 111808-06
GENERAL CONDITIONS
Cross Section: HOMO NANO
Pack Build: 1960 100 HPI
Polymer B: 1800 PP Drying Time: n/a
Gm Hole Min: 0.0057
Run Speed: 2.1 mpm
Meter Pump Size: 2.92 cc/rev
Melt Pump Speed 3.8 RPM
Aspirator Pres: 3.75
Forming Table Distance: 8 in
Room Temp:
Air heat set @ 335C, actual 267C.
No Scrim
185
EXTRUSION TEMPERATURES SAMPLE 111808-06
DEVICE Measured Value oC Setpoint oC
1. EXTRUDER A ZONE 1 TMP 180.0 180.0 2. EXTRUDER A ZONE 2 TMP 180.0 180.0 3. EXTRUDER A ZONE 3 TMP 180.0 180.0 4. EXTRUDER A ZONE 4 TMP 180.0 180.0 5. EXTRUDER A FLANGE TMP 200.0 200.0 6. EXTRUDER B ZONE 1 TMP 180.0 180.0 7. EXTRUDER B ZONE 2 TMP 179.0 180.0 8. EXTRUDER B ZONE 3 TMP 180.0 180.0 9. EXTRUDER B ZONE 4 TMP 180.0 180.0 10. EXTRUDER B FLANGE TMP 200.0 200.0 11. Inlet manifold flange TMP 269.0 - 12. MELT PUMP B HEATER - - 13. TRANSFER LINE A ZONE 1 TMP 200.0 200.0 14. TRANSFER LINE A ZONE 2 TMP 200.0 200.0 15. TRANSFER LINE A ZONE 3 TMP 201.0 200.0 16. TRANSFER LINE A ZONE 4 TMP 200.0 200.0 17. TRANSFER LINE B ZONE 1 TMP 230.0 230.0 18. TRANSFER LINE B ZONE 2 TMP 230.0 230.0 19. TRANSFER LINE B ZONE 3 TMP 230.0 230.0 20. TRANSFER LINE B ZONE 4 TMP 240.0 240.0 21. REGEN HEATER 19.0 - 22. SPIN BEAM ZONE 1 TMP 246.0 245.0 23. SPIN BEAM ZONE 2 TMP 245.0 245.0 24. SPIN BEAM ZONE 3 TMP 245.0 245.0 25. SPIN BEAM ZONE 4 TMP 245.0 245.0 26. SPIN BEAM ZONE 5 TMP 245.0 245.0 27. SPIN BEAM ZONE 6 TMP 245.0 245.0 28. SPIN BEAM ZONE 7 TMP 246.0 245.0 29. SPIN BEAM ZONE 8 TMP 245.0 245.0 30. SPIN BEAM AIR TMP - - 31. A TRANSFER LINE OUTLET TMP 188.0 - 32. B TRANSFER LINE OUTLET TMP 211.0 - 33. A MELT PUMP OUTLET TMP 24.0 - 34. B MELT PUMP OUTLET TMP 238.0 - 35. QUENCH AIR 1 TMP 22.0 - 36. INTERNAL PACK TEMP 244.0 - 37. AIR MANIFOLD IN TEMP 280.0 - 38. QUENCH AIR CONTROL TMP 24.0 5.0 39. PACK PREHEATER CONTROL TMP22.0 -
186
SAMPLE 111808-07
GENERAL CONDITIONS
Cross Section: HOMO NANO
Pack Build: 1960 100 HPI
Polymer B: 1800 PP
Gm Hole Min: 0.0055
Run Speed: 4.9 mpm
Meter Pump Size: 2.92 cc/rev
Melt Pump Speed 5.0 RPM
Aspirator Pres: 5
Forming Table Distance: 8 in
Room Temp:
Air heat set @ 340C, actual 283C.
Polyethylene Scrim
187
EXTRUSION TEMPERATURES SAMPLE 111808-07
DEVICE Measured Value oC Setpoint oC
1. EXTRUDER A ZONE 1 TMP 180.0 180.0 2. EXTRUDER A ZONE 2 TMP 180.0 180.0 3. EXTRUDER A ZONE 3 TMP 180.0 180.0 4. EXTRUDER A ZONE 4 TMP 180.0 180.0 5. EXTRUDER A FLANGE TMP 200.0 200.0 6. EXTRUDER B ZONE 1 TMP 180.0 180.0 7. EXTRUDER B ZONE 2 TMP 180.0 180.0 8. EXTRUDER B ZONE 3 TMP 180.0 180.0 9. EXTRUDER B ZONE 4 TMP 180.0 180.0 10. EXTRUDER B FLANGE TMP 200.0 200.0 11. Inlet manifold flange TMP 285.0 - 12. MELT PUMP B HEATER - - 13. TRANSFER LINE A ZONE 1 TMP 200.0 200.0 14. TRANSFER LINE A ZONE 2 TMP 200.0 200.0 15. TRANSFER LINE A ZONE 3 TMP 201.0 200.0 16. TRANSFER LINE A ZONE 4 TMP 200.0 200.0 17. TRANSFER LINE B ZONE 1 TMP 230.0 230.0 18. TRANSFER LINE B ZONE 2 TMP 230.0 230.0 19. TRANSFER LINE B ZONE 3 TMP 235.0 235.0 20. TRANSFER LINE B ZONE 4 TMP 250.0 250.0 21. REGEN HEATER 20.0 - 22. SPIN BEAM ZONE 1 TMP 255.0 255.0 23. SPIN BEAM ZONE 2 TMP 255.0 255.0 24. SPIN BEAM ZONE 3 TMP 255.0 255.0 25. SPIN BEAM ZONE 4 TMP 255.0 255.0 26. SPIN BEAM ZONE 5 TMP 255.0 255.0 27. SPIN BEAM ZONE 6 TMP 255.0 255.0 28. SPIN BEAM ZONE 7 TMP 255.0 255.0 29. SPIN BEAM ZONE 8 TMP 255.0 255.0 30. SPIN BEAM AIR TMP - - 31. A TRANSFER LINE OUTLET TMP 190.0 - 32. B TRANSFER LINE OUTLET TMP 219.0 - 33. A MELT PUMP OUTLET TMP 20.0 15.0 34. B MELT PUMP OUTLET TMP 246.0 - 35. QUENCH AIR 1 TMP 23.0 - 36. INTERNAL PACK TEMP 257.0 - 37. AIR MANIFOLD IN TEMP 295.0 - 38. QUENCH AIR CONTROL TMP 24.0 5.0 39. PACK PREHEATER CONTROL TMP - -
188
SAMPLE 111808-08
GENERAL CONDITIONS
Cross Section: HOMO NANO
Pack Build: 1960 100 HPI
Polymer B: 1800 PP
Gm Hole Min: 0.0055
Run Speed: 9.6 mpm
Meter Pump Size: 2.92 cc/rev
Melt Pump Speed 10.9 RPM
Aspirator Pres: 5
Forming Table Distance: 8 in
Room Temp:
Air heat set @ 340C, actual 283C.
Polyethylene Scrim
189
EXTRUSION TEMPERATURES SAMPLE 111808-08
DEVICE Measured Value oC Setpoint oC
1. EXTRUDER A ZONE 1 TMP 181.0 180.0 2. EXTRUDER A ZONE 2 TMP 181.0 180.0 3. EXTRUDER A ZONE 3 TMP 180.0 180.0 4. EXTRUDER A ZONE 4 TMP 180.0 180.0 5. EXTRUDER A FLANGE TMP 200.0 200.0 6. EXTRUDER B ZONE 1 TMP 180.0 180.0 7. EXTRUDER B ZONE 2 TMP 180.0 180.0 8. EXTRUDER B ZONE 3 TMP 180.0 180.0 9. EXTRUDER B ZONE 4 TMP 180.0 180.0 10. EXTRUDER B FLANGE TMP 200.0 200.0 11. Inlet manifold flange TMP 286.0 - 12. MELT PUMP B HEATER - - 13. TRANSFER LINE A ZONE 1 TMP 200.0 200.0 14. TRANSFER LINE A ZONE 2 TMP 200.0 200.0 15. TRANSFER LINE A ZONE 3 TMP 200.0 200.0 16. TRANSFER LINE A ZONE 4 TMP 200.0 200.0 17. TRANSFER LINE B ZONE 1 TMP 230.0 230.0 18. TRANSFER LINE B ZONE 2 TMP 230.0 230.0 19. TRANSFER LINE B ZONE 3 TMP 235.0 235.0 20. TRANSFER LINE B ZONE 4 TMP 250.0 250.0 21. REGEN HEATER 20.0 - 22. SPIN BEAM ZONE 1 TMP 256.0 255.0 23. SPIN BEAM ZONE 2 TMP 255.0 255.0 24. SPIN BEAM ZONE 3 TMP 255.0 255.0 25. SPIN BEAM ZONE 4 TMP 255.0 255.0 26. SPIN BEAM ZONE 5 TMP 255.0 255.0 27. SPIN BEAM ZONE 6 TMP 255.0 255.0 28. SPIN BEAM ZONE 7 TMP 255.0 255.0 29. SPIN BEAM ZONE 8 TMP 255.0 255.0 30. SPIN BEAM AIR TMP - - 31. A TRANSFER LINE OUTLET TMP 190.0 - 32. B TRANSFER LINE OUTLET TMP 220.0 - 33. A MELT PUMP OUTLET TMP 25.0 - 34. B MELT PUMP OUTLET TMP 248.0 - 35. QUENCH AIR 1 TMP 23.0 - 36. INTERNAL PACK TEMP 258.0 - 37. AIR MANIFOLD IN TEMP 296.0 - 38. QUENCH AIR CONTROL TMP 25.0 5.0 39. PACK PREHEATER CONTROL TMP23.0 -
190
SAMPLE 111808-09
GENERAL CONDITIONS
Cross Section: HOMO NANO
Pack Build: 1960 100 HPI
Polymer B: 1800 PP
Gm Hole Min: 0.0055
Run Speed: 19.62 mpm
Meter Pump Size: 2.92 cc/rev
Melt Pump Speed 5.0 RPM
Aspirator Pres: 5
Forming Table Distance: 8 in
Room Temp:
Air heat set @ 340C, actual 283C.
Polyethylene Scrim
191
EXTRUSION TEMPERATURES SAMPLE 111808-09
DEVICE Measured Value oC Setpoint oC
1. EXTRUDER A ZONE 1 TMP 182.0 180.0 2. EXTRUDER A ZONE 2 TMP 180.0 180.0 3. EXTRUDER A ZONE 3 TMP 180.0 180.0 4. EXTRUDER A ZONE 4 TMP 180.0 180.0 5. EXTRUDER A FLANGE TMP 200.0 200.0 6. EXTRUDER B ZONE 1 TMP 179.0 180.0 7. EXTRUDER B ZONE 2 TMP 181.0 180.0 8. EXTRUDER B ZONE 3 TMP 180.0 180.0 9. EXTRUDER B ZONE 4 TMP 180.0 180.0 10. EXTRUDER B FLANGE TMP 200.0 200.0 11. Inlet manifold flange TMP 286.0 - 12. MELT PUMP B HEATER - - 13. TRANSFER LINE A ZONE 1 TMP 200.0 200.0 14. TRANSFER LINE A ZONE 2 TMP 200.0 200.0 15. TRANSFER LINE A ZONE 3 TMP 200.0 200.0 16. TRANSFER LINE A ZONE 4 TMP 200.0 200.0 17. TRANSFER LINE B ZONE 1 TMP 230.0 230.0 18. TRANSFER LINE B ZONE 2 TMP 230.0 230.0 19. TRANSFER LINE B ZONE 3 TMP 235.0 235.0 20. TRANSFER LINE B ZONE 4 TMP 250.0 250.0 21. REGEN HEATER 20.0 - 22. SPIN BEAM ZONE 1 TMP 256.0 255.0 23. SPIN BEAM ZONE 2 TMP 255.0 255.0 24. SPIN BEAM ZONE 3 TMP 255.0 255.0 25. SPIN BEAM ZONE 4 TMP 255.0 255.0 26. SPIN BEAM ZONE 5 TMP 255.0 255.0 27. SPIN BEAM ZONE 6 TMP 255.0 255.0 28. SPIN BEAM ZONE 7 TMP 255.0 255.0 29. SPIN BEAM ZONE 8 TMP 255.0 255.0 30. SPIN BEAM AIR TMP - - 31. A TRANSFER LINE OUTLET TMP 190.0 - 32. B TRANSFER LINE OUTLET TMP 220.0 - 33. A MELT PUMP OUTLET TMP 25.0 - 34. B MELT PUMP OUTLET TMP 249.0 - 35. QUENCH AIR 1 TMP 22.0 - 36. INTERNAL PACK TEMP 258.0 - 37. AIR MANIFOLD IN TEMP 296.0 - 38. QUENCH AIR CONTROL TMP 25.0 5.0 39. PACK PREHEATER CONTROL TMP23.0 -
192
SAMPLE 111808-10
GENERAL CONDITIONS
Cross Section: HOMO NANO
Pack Build: 1960 100 HPI
Polymer B: 1800 PP
Gm Hole Min: 0.0055
Run Speed: 6.8 mpm
Meter Pump Size: 2.92 cc/rev
Melt Pump Speed 5.0 RPM
Aspirator Pres: 5
Forming Table Distance: 8 in
Room Temp:
Air heat set @ 350C, actual 254C.
Polyethylene Scrim
193
EXTRUSION TEMPERATURES SAMPLE 111808-10
DEVICE Measured Value oC Setpoint oC
1. EXTRUDER A ZONE 1 TMP 181.0 180.0 2. EXTRUDER A ZONE 2 TMP 180.0 180.0 3. EXTRUDER A ZONE 3 TMP 180.0 180.0 4. EXTRUDER A ZONE 4 TMP 180.0 180.0 5. EXTRUDER A FLANGE TMP 200.0 200.0 6. EXTRUDER B ZONE 1 TMP 180.0 180.0 7. EXTRUDER B ZONE 2 TMP 180.0 180.0 8. EXTRUDER B ZONE 3 TMP 180.0 180.0 9. EXTRUDER B ZONE 4 TMP 180.0 180.0 10. EXTRUDER B FLANGE TMP 200.0 200.0 11. Inlet manifold flange TMP 282.0 - 12. MELT PUMP B HEATER - - 13. TRANSFER LINE A ZONE 1 TMP 200.0 200.0 14. TRANSFER LINE A ZONE 2 TMP 200.0 200.0 15. TRANSFER LINE A ZONE 3 TMP 202.0 200.0 16. TRANSFER LINE A ZONE 4 TMP 200.0 200.0 17. TRANSFER LINE B ZONE 1 TMP 230.0 230.0 18. TRANSFER LINE B ZONE 2 TMP 230.0 230.0 19. TRANSFER LINE B ZONE 3 TMP 235.0 235.0 20. TRANSFER LINE B ZONE 4 TMP 250.0 250.0 21. REGEN HEATER 20.0 - 22. SPIN BEAM ZONE 1 TMP 255.0 255.0 23. SPIN BEAM ZONE 2 TMP 255.0 255.0 24. SPIN BEAM ZONE 3 TMP 255.0 255.0 25. SPIN BEAM ZONE 4 TMP 255.0 255.0 26. SPIN BEAM ZONE 5 TMP 255.0 255.0 27. SPIN BEAM ZONE 6 TMP 255.0 255.0 28. SPIN BEAM ZONE 7 TMP 255.0 255.0 29. SPIN BEAM ZONE 8 TMP 255.0 255.0 30. SPIN BEAM AIR TMP - - 31. A TRANSFER LINE OUTLET TMP 191.0 - 32. B TRANSFER LINE OUTLET TMP 220.0 - 33. A MELT PUMP OUTLET TMP 25.0 - 34. B MELT PUMP OUTLET TMP 250.0 - 35. QUENCH AIR 1 TMP 22.0 - 36. INTERNAL PACK TEMP 254.0 - 37. AIR MANIFOLD IN TEMP 298.0 - 38. QUENCH AIR CONTROL TMP 24.0 5.0 39. PACK PREHEATER CONTROL TMP23.0 -
194
SAMPLE 111808-11
GENERAL CONDITIONS
Cross Section: HOMO NANO
Pack Build: 1960 100 HPI
Polymer B: 1800 PP
Gm Hole Min: 0.0023
Run Speed: 6.8 mpm
Meter Pump Size: 2.92 cc/rev
Melt Pump Speed 2.1 RPM
Aspirator Pres: 5
Forming Table Distance: 8 in
Room Temp:
Air heat set @ 360C, actual 284C.
Polyethylene Scrim
195
EXTRUSION TEMPERATURES SAMPLE 111808-11
DEVICE Measured Value oC Setpoint oC
1. EXTRUDER A ZONE 1 TMP 181.0 180.0 2. EXTRUDER A ZONE 2 TMP 180.0 180.0 3. EXTRUDER A ZONE 3 TMP 180.0 180.0 4. EXTRUDER A ZONE 4 TMP 180.0 180.0 5. EXTRUDER A FLANGE TMP 200.0 200.0 6. EXTRUDER B ZONE 1 TMP 180.0 180.0 7. EXTRUDER B ZONE 2 TMP 180.0 180.0 8. EXTRUDER B ZONE 3 TMP 181.0 180.0 9. EXTRUDER B ZONE 4 TMP 180.0 180.0 10. EXTRUDER B FLANGE TMP 200.0 200.0 11. Inlet manifold flange TMP 288.0 - 12. MELT PUMP B HEATER - - 13. TRANSFER LINE A ZONE 1 TMP 200.0 200.0 14. TRANSFER LINE A ZONE 2 TMP 200.0 200.0 15. TRANSFER LINE A ZONE 3 TMP 200.0 200.0 16. TRANSFER LINE A ZONE 4 TMP 200.0 200.0 17. TRANSFER LINE B ZONE 1 TMP 230.0 230.0 18. TRANSFER LINE B ZONE 2 TMP 230.0 230.0 19. TRANSFER LINE B ZONE 3 TMP 235.0 235.0 20. TRANSFER LINE B ZONE 4 TMP 250.0 250.0 21. REGEN HEATER 20.0 - 22. SPIN BEAM ZONE 1 TMP 255.0 255.0 23. SPIN BEAM ZONE 2 TMP 255.0 255.0 24. SPIN BEAM ZONE 3 TMP 255.0 255.0 25. SPIN BEAM ZONE 4 TMP 255.0 255.0 26. SPIN BEAM ZONE 5 TMP 255.0 255.0 27. SPIN BEAM ZONE 6 TMP 255.0 255.0 28. SPIN BEAM ZONE 7 TMP 255.0 255.0 29. SPIN BEAM ZONE 8 TMP 255.0 255.0 30. SPIN BEAM AIR TMP - - 31. A TRANSFER LINE OUTLET TMP 191.0 - 32. B TRANSFER LINE OUTLET TMP 216.0 - 33. A MELT PUMP OUTLET TMP 258.0 - 34. B MELT PUMP OUTLET TMP 252.0 - 35. QUENCH AIR 1 TMP 23.0 - 36. INTERNAL PACK TEMP 255.0 - 37. AIR MANIFOLD IN TEMP 304.0 - 38. QUENCH AIR CONTROL TMP 25.0 5.0 39. PACK PREHEATER CONTROL TMP 23.0 -
196
SAMPLE 111808-12
GENERAL CONDITIONS
Cross Section: HOMO NANO
Pack Build: 1960 100 HPI
Polymer B: 1800 PP
Gm Hole Min: 0.0023
Run Speed: 4.48 mpm
Meter Pump Size: 2.92 cc/rev
Melt Pump Speed 2.1 RPM
Aspirator Pres: 4
Forming Table Distance: 8 in
Room Temp:
Air heat set @ 360C, actual 288C.
Polyethylene Scrim
197
EXTRUSION TEMPERATURES SAMPLE 111808-12
DEVICE Measured Value oC Setpoint oC
1. EXTRUDER A ZONE 1 TMP 181.0 180.0 2. EXTRUDER A ZONE 2 TMP 180.0 180.0 3. EXTRUDER A ZONE 3 TMP 180.0 180.0 4. EXTRUDER A ZONE 4 TMP 180.0 180.0 5. EXTRUDER A FLANGE TMP 200.0 200.0 6. EXTRUDER B ZONE 1 TMP 180.0 180.0 7. EXTRUDER B ZONE 2 TMP 179.0 180.0 8. EXTRUDER B ZONE 3 TMP 180.0 180.0 9. EXTRUDER B ZONE 4 TMP 180.0 180.0 10. EXTRUDER B FLANGE TMP 200.0 200.0 11. Inlet manifold flange TMP 291.0 - 12. MELT PUMP B HEATER - - 13. TRANSFER LINE A ZONE 1 TMP 200.0 200.0 14. TRANSFER LINE A ZONE 2 TMP 200.0 200.0 15. TRANSFER LINE A ZONE 3 TMP 200.0 200.0 16. TRANSFER LINE A ZONE 4 TMP 200.0 200.0 17. TRANSFER LINE B ZONE 1 TMP 230.0 230.0 18. TRANSFER LINE B ZONE 2 TMP 230.0 230.0 19. TRANSFER LINE B ZONE 3 TMP 235.0 235.0 20. TRANSFER LINE B ZONE 4 TMP 250.0 250.0 21. REGEN HEATER 20.0 - 22. SPIN BEAM ZONE 1 TMP 255.0 255.0 23. SPIN BEAM ZONE 2 TMP 255.0 255.0 24. SPIN BEAM ZONE 3 TMP 255.0 255.0 25. SPIN BEAM ZONE 4 TMP 255.0 255.0 26. SPIN BEAM ZONE 5 TMP 255.0 255.0 27. SPIN BEAM ZONE 6 TMP 255.0 255.0 28. SPIN BEAM ZONE 7 TMP 255.0 255.0 29. SPIN BEAM ZONE 8 TMP 255.0 255.0 30. SPIN BEAM AIR TMP - - 31. A TRANSFER LINE OUTLET TMP 191.0 - 32. B TRANSFER LINE OUTLET TMP 215.0 - 33. A MELT PUMP OUTLET TMP 258.0 - 34. B MELT PUMP OUTLET TMP 253.0 - 35. QUENCH AIR 1 TMP 22.0 - 36. INTERNAL PACK TEMP 256.0 - 37. AIR MANIFOLD IN TEMP 307.0 - 38. QUENCH AIR CONTROL TMP 24.0 5.0 39. PACK PREHEATER CONTROL TMP23.0 -
198
References
(1) Davis, G. B. HEPA room air purifier, Patent 4,629,482, 1986
(2) Rathey, A. HEPA Help, American School & University
(3) Walters, D., The ins and outs of HEPA filters, Laboratory Health & Safety Division of Chemical Health and Safety of the American Chemical Society
(4) Thomas, D.; Contal, P.; Renaudin, V.; Penicot, P.; Leclerc, D.; Vendel, J., Modeling pressure drop in HEPA filters during dynamic filtration, J. Aerosol Sci. 1999, 30, 235-246
(5) Healy, J. F. In Natural History: A selection by Pliny (the elder); Penguin Group: New York, NY, 1991; pp 304
(6) Spurny, K. R., Ed.; In Advances in Aerosol Filtration Science; CRC Press: Boca Raton, FL, USA, 1997
(7) The History of Air Purifiers. http://www.achooallergy.com/airpurifiers-history.asp
(8) Allergy Relief through Allergen Avoidance. http://www.achooallergy.com/allergy-relief-learning-center.asp
(9) Schnarr, B. The History of Air Filters - Breath Easy! http://www.thehistoryof.net/history-of-air-filters.html (accessed 01/2010)
(10) E. Littell, R. S. L., The Living Age, 1876, Fifth Series, Volume 8, 701
(11) Gilbert, H. High-efficiency Particulate Air Filter Units, Inspection, Handling, Installation, AEC Report 1961, TID-7023, NTIS
(12) Antosia, E.; Cahill, J. D., Eds.; In Bioterroism & Disaster Medicine; Springer Science & Business Media LLC: New York, NY, 2006; pp 221
(13) Hutten, I. In Handbook of Nonwoven Filter Media; Elsevier Ltd: 2007
(14) Virta, R. L. Some facts about asbestos, U.S. Geological Survey: 2001
(15) Van Turnhout, J.; Hoeneveld, W.; Adamse, J.; Van Rossen, L,. Electret Filters for High-Efficiency and High-Flow Air Cleaning, IEEE TRANS.INDUST.APPLIC. 1981, 17, 240-248
(16) Carlsson, T., Indoor air filtration: Why use polymer based filter media? Filtration Sep. 2001, 38, 30-32
199
(17) Adamiak, K., Viscous flow model for charged particle trajectories around a single square fiber in an electric field, IEEE Trans. Ind. Appl. 1999, 35, 352-358
(18) Cox, C. L.; Brown, P. J.; Larzelere, J. C. Simulation of C-CP Fiber-Based Air Filtration, Journal of Engineered Fibers and Fabrics SPECIAL ISSUE 2008 – FILTRATION
(19) Cox, C. L.; Brown, P. J.; Larzelere, J. C. Modeling the performance of C-CP fiber-based HEPA filter, AATCC Manuscript
(20) Pich, J. Teorie filtrace aerosolů vláknitými a membránovými filtry, 1964
(21) Wang, J.; Kim, S. C.; Pui, D. Y. H. J., Investigation of the figure of merit for filters with a single nanofiber layer on a substrate, Aerosol Sci. 2008, 39, 323-334
(22) Heikkilä, P., Nanostructured Fibre Composites, and Materials for Air Filtration, ISSN/ISBN 978-952-15-2023-5
(23) Graham, K. M.; Quyang, M.; Raether T.; Grafe, T. H.; McDonalds, B.; Koehl, L., Polymeric nanofibers in air filtration applications, Advancements in Filtration and Separation Technology 2002, 25
(24) Hinds, W. C. In Aerosol Technology, Properties Behaviour & Measurements of Airborne Particles; John Wiley & Sons: New York, 1999; pp 483
(25) Grafe, T. H.; Graham, K. M. In In Nanofiber webs from electrospinning; Proceedings of the 5th International Conference on Nonwovens in Filtration; pp 1–5
(26) Vaughn, E.; Ramachandran, G., Fiberglass vs. synthetic air filtration media, INJ Fall 2002, 41-53
(27) Reneker, D. H.; Fong, H. In Polymeric Nanofibers; Oxford University Press: 2006
(28) Formhals, A., Process and apparatus for preparing artificial threads, Patent 1,975,504, 1934
(29) Ahn, Y.; Park, S.; Kim, G.; Hwang, Y.; Lee, C.; Shin, H.; Lee, J., Development of high efficiency nanofilters made of nanofibers Curr.Appl.Phys, 2006, 6, 1030–1035
(30) Kim, S. J.; Nam, Y. S.; Rhee, D. M.; Park, H. S.; Park, W. H., Preparation and characterization of antimicrobial polycarbonate nanofibrous membrane, European polymer journal 2007, 43, 3146-3152
(31) Reneker, D. H.; Chun, I., Nanometre diameter fibres of polymer, produced by electrospinning, Nanotechnology, 1996, 7
200
(32) Lyons, J.; Kaufmann, J. Electrospinning Polymers from Molten State, http://www.novacompinc.com/papers/Molten.html
(33) Pawlowski, K. J.; Barnes, C. P.; Boland, E. D.; Wnek, G. E.; Bowlin, G. L., Biomedical Nanoscience: Electrospinning Basic Concepts, Applications, and Classroom Demonstration, JOURNAL OF MATERIALS EDUCATION 2004, 26, 195
(34) Brown, P. J.; Stevens, K., Eds.; In Nanofibers and nanotechnology in textiles; Woodhead Publishing Limited and CRC Press LLC: Boca Raton, FL, 2007; pp 528
(35) Shipp Jr. P.W.; Vogt, C.M. Melt-blown material with depth fiber size gradient, Patent 4,714,647, 1987
(36) Dugan, J.; Homonoff, E.; Mechanisms, F., Synthetic Split Microfiber Technology for Filtration INDA Filtration, 28-30
(37) Kulshreshtha, A. K.; Vasile, C. In Handbook of polymer composites, Volume 3; Rapra Technology Limited: 2003
(38) Raghavendra R.H.; Hege, R.; Dahiya, A.; Kamath, M. G. Nanofiber Nonwovens. http://www.engr.utk.edu/mse/pages/Textiles/Nanofiber%20Nonwovens.htm
(39) Hills Inc. http://www.hillsinc.net/submicron%20.shtml (accessed October, 19, 2009)
(40) Nakajima, T.; Kajiwara, K.; McIntyre, J. E., Eds.; In Advanced fiber spinning technology; Woodhead Publishing Limited: Cambridge, England, 2000
(41) Toshinari, N.; Kajiwara, K.; McIntyre, J. E. In Advanced fiber spinning technologies; Woodhead Publishing Ltd: Abington, Cambridge, 2000
(42) Reavill, G. In Aftermath Inc.: Cleaning up after CSI; Penguin Group (USA) Inc.: New York, NY, 2007
(43) Sherman, M., Ed.; In Medical device packaging handbook; Marcel Dekker Inc.: New York, NY, 1998
(44) Shin, H.; Siemionko, R. K., Flash spinning process, Patent 5672307, 1997
(45) Coates, D. M.; McMillin, C. K.; Chen, J. C, Process for flash spinning fiber-forming polymers Patent 5192468, 1993
(46) Gregor, E. C. Primer on Nonwoven Fabric Filtration Media, http://www.egregor.com/whitepapers/NONWOVEN%20PRIMER.pdf (accessed Sept. 8, 2008)
201
(47) Wang, X.; Gong, R. J., Thermally bonded nonwoven filters composed of bicomponent polypropylene/polyester fiber. I. Statistical approach for minimizing the pore size, Appl Polym Sci 2006, 101, 2689-2699
(48) Jaganathan, S.; Vahedi Tafreshi, H.; Pourdeyhimi, B., A realistic approach for modeling permeability of fibrous media: 3-D imaging coupled with CFD simulation A realistic approach for modeling permeability of fibrous media: 3-D imaging coupled with CFD simulation,Chemical Engineering Science 2008, 63, 244-252
(49) Kadolph, S. J, Langford, A. L. In Textiles; Pearson Education, Inc.: Uppser Saddle River, New Jersey, 07458, 2002
(50) Mahall, K. In Quality of Assessment of Textiles; Springer: Germany, 2003; pp 37-42
(51) Bueno, M.; Aneja, A.; Renner, M., Influence of the shape of fiber cross section on fabric surface characteristics J. Mater. Sci. 2004, 39, 557-564
(52) Novick, V.; Monson, P.; Ellison, P., The effect of solid particle mass loading on the pressure drop of HEPA filters, J. Aerosol Sci. 1992, 23, 657-665
(53) Rohrbach et al., Complex shaped fiber for particle and molecular filtration, US Patent 7,442,223 B2, 2008
(54) Zhu, H.; Hinestroza, J. P., Collection efficiency for filters with staggered parallel Y and triple Y fibers: A numerical study, Journal of Engineered Fibers and Fabrics , 4, 16
(55) Phillips, B. M.; Bagrodia, S., Process of making fibers, Patent 5,611,981, 1997
(56) Phillips, B. M.; Wilson, S. A.; Pollock, M. A., Tobacco smoke filter, Patent 5,356,704, 1994
(57) Phillips, B. M.; Nelson, J. L., Insulation material, Patent 5,837,625, 1998
(58) Paul, H. L.; Diller, K. R., Comparison of thermal insulation performance of fibrous materials for the advanced space suit, Journal of Biomedical Engineering Oct, 2003, 125, 37
(59) Fuller, L. G., Production and characterization of novel shaped fibers for fluid adsorption and transport, Clemson University, 2009
(60) Lydall Filtration/Separation. http://www.lydalfiltration.com/products/air/mg.shtml
(61) Keeler, S.; Cambo, W. H., Composite Dual Layer HEPA Filter, US Patent 2007/0163218, 2006
202
(62) Jena, A.; Gupta, K., Fast evaluation of average fiber diameters of nonwovens, Porous Materials, Inc .
(63) Lawrence, C. A.; Pinghui, L., Relation of structure, properties and performance of fibrous media for gas filtration, Chem. Eng. Technol. 2006, 29, 957
(64) Podgórski, A.; Bałazy, A.; Gradoń, L., Application of nanofibers to improve the filtration efficiency of the most penetrating aerosol particles in fibrous filters, Chemical Engineering Science 2006, 61, 6804-6815
(65) Quevedo, J.; Patel, G.; Pfeffer, R.; Dave, R., Agglomerates and granules of nanoparticles as filter media for submicron particles, Powder Technol 2008, 183, 480-500
(66) Kim, G. T.; Ahn, Y. C.; Lee, J. K., Characteristics of Nylon 6 nanofilter for removing ultra fine particles, Korean Journal of Chemical Engineering 2008, 25, 368-372
(67) Lee, K. M., R., Filter Collection, Aerosol measurement: Principles, techniques, and applications 2001, 197
(68) Kuraray American, Inc. 2010, http://www.kuraray-am.com/
(69) Lau, W.S., Infrared characterization for microelectronics, World Scientific Publishing, River Edge, NJ 07661, 1999
(70) Noda, I., Dowrey, A.E., Haynes, J.L., Marcott, C., Group frequency assignments for major infrared bands observed in common synthetic polymers, The Proctor & Gamble Company, Beckett Technical Center, Chapter 22, 395
(71) Buckingham, P., Simulation of capillary-channeled polymer fiber-based particulate air filtration, Clemson University, 2010
(72) Coury, J. S. J. R., Collection efficiency of fiber filters operating on the removal of nano-sized aerosol particles II. Heterogeneous fibers, Separation and Purification Technology 2007, 58
(73) Dimassi, M.; Koehl, L.; Zeng, X.; Peruwelz, A., Pore network modelling using image processing techniques, Science and Technology 2008, 20, 137-149
(74) Faley, T. L.; Strieder, W., Knudsen flow through a random bed of unidirectional fibers, J. Appl. Phys. 1987, 62, 4394
(75) George, N. A.; Sutsko, M. G.; McKenna, D. B., ULPA Filter, Patent 5507847, 1996
203
(76) Heim, M.; Mullins, B. J.; Wild, M.; Meyer, J.; Kasper, G., Filtration efficiency of aerosol particles below 20 nanometers, Aerosol Science and Technology 2005, 39, 782-789
(77) Hou, D.; Maheshwari, S.; Chang, H. C., Rapid bioparticle concentration and detection by combining a discharge driven vortex with surface enhanced Raman scattering, Biomicrofluidics 2007, 1, 014106
(78) Jackson, G. W.; James, D. F., The permeability of fibrous porous media,Can. J. Chem. Eng. 1986, 64, 364-374
(79) Japuntich, D. A.; Stenhouse, J. I. T.; Liu, B. Y. H., Effective pore diameter and monodisperse particle clogging of fibrous filters, J. Aerosol Sci. 1997, 28, 147
(80) Kikutani, I.; Radhakrishnan, J.; Arikawa, S.; Takaku, A.; Okui, N.; Jin, N.; Niwa, F.; Kudo, Y., High-Speed Melt Spinning of Bicomponent Fibers: Mechanism of Fiber Structure Development in Poly (ethylene terephtalate)/Propylene System, J. Appl. Pol. Sci 1996, 62, 1913
(81) Kim, S. C.; Harrington, M. S.; Pui, D. Y. H., Experimental study of nanoparticles penetration through commercial filter media, Journal of Nanoparticle Research 2007, 9, 117-125
(82) Labrecque, R. P., The effects of fiber cross-sectional shape on the resistance to the flow of fluids through fiber mats, Tappi 1968, 51, 8
(83) Lahbabi, A.; Chang, H. C., Flow in periodically constricted tubes: transition to inertial and nonsteady flows, Chemical engineering science 1986, 41, 2487
(84) Lahbabi, A.; Chang, H. C., High Reynolds number flow through cubic arrays of spheres: steady-state solution and transition to turbulence, Chemical engineering science 1985, 40, 435-447
(85) Lawrence, C. A.; Pinghui, L., Relation of structure, properties and performance of fibrous media for gas filtration, Chem. Eng. Technol. 2006, 29, 957
(86) Lee, K. M., R. Aerosol measurement: Principles, techniques, and applications 2001, 197.
(87) Lee, K.; Liu, B., Filter Collection, Aerosol Science and Technology 1982, 1, 147-161
(88) Maze, B.; Tafreshi, H. V.; Pourdeyhimi, B., Case studies of air filtration at microscales: Micro- and Nanofiber media, Journal of Engineered Fibers and Fabrics 2008
(89) Moelter, W.; Fissan, H., Structure of a high efficiency glass fiber filter medium, Aerosol Science and Technology 1997, 27, 447
204
(90) Morris, W. J., Textiles, Fibres, Yarns, and Fabrics for Gas Filtration, Fibrous and related materials for the filtration of gases, Shirley Institute, Publication S27
(91) Natale, A. Disposable HEPA filtration device, Patent 4726825, 1986
(92) Nilsson, L.; Stenston, S., Gas diffusion through sheets of fibrous porous media, Chemical Engineering Science 1995, 50, 361-371
(93) Oh, Y.; Jeon, K.; Jung, A.; Jung, Y., A simulation study on the collection of submicron particles in a unipolar charged fiber, Aerosol Science and Technology 2002, 36, 573-582
(94) Pollard, W.; Present, R., On gaseous self-diffusion in long capillary tubes, Phys.Rev Phys Rev 1946, 73, 762
(95) Pourdeyhimi, B. Comments on the Paper Entitled ―Splitting of Islands-in-the-Sea Fibers (PA6/COPET) During Hydroentangling of Nonwovens‖
(96) Song, C. B.; Lee, J. L.; Park, H. S.; Lee, K. W., Effect of solid monodisperse particles on the pressure drop of fibrous filters, Korean Journal of Chemical Engineering 2007, 24, 148-153
(97) Steffens, J.; Coury, J., Collection efficiency of fiber filters operating on the removal of nano-sized aerosol particles: I—Homogeneous fibers, Separation and Purification Technology 2007, 58, 99-105.
(98) Teo, W.; Ramakrishna, S., A review on electrospinning design and nanofibre assemblies, Nanotechnology 2006, 17, R89-R106.
(99) Termonia, Y., Permeability of sheets of nonwoven fibrous media, Chemical Engineering Science 1998, 53, 1203-1208.
(100) Ullrich, A.; Coussens, L.; Hayflick, J.; Dull, T.; Gray, A.; Tam, A.; Lee, J.; Yarden, Y.; Libermann, T.; Schlessinger,, Human epidermal growth factor receptor cDNA sequence and aberrant expression of the amplified gene in A431 epidermoid carcinoma cells, Nature Publishing Group, 1984
(101) Walsh, D.; Stenhouse, J., The effect of particle size, charge, and composition on the loading characteristics of an electrically active fibrous filter material, J. Aerosol Sci. 1997, 28, 307-321
(102) Wang, J.; Pui, D. Y. H., Filtration of aerosol particles by elliptical fibers: a numerical study, Journal of Nanoparticle Research 2009, 11, 185-196
205
(103) Wang, X.; Gong, R., Thermally bonded nonwoven filters composed of bi-component polypropylene/polyester fiber. II. Relationships between fabric area density, air permeability, and pore size distribution, J Appl Polym Sci 2006, 102, 2264-2275
(104) Wu, J.; Ben, Y.; Battigelli, D.; Chang, H. C., Long-range AC electroosmotic trapping and detection of bioparticles, Ind Eng Chem Res 2005, 44, 2815-2822
(105) Yasuda, H.; Tsai, J. T., Pore size of microporous polymer membranes, Journal of Applied Polymer Science 1974, 18, 805
(106) Zhang, Y.; Wang, H.; Zhang, C.; Chen, Y., Modeling of capillary flow in shaped polymer fiber bundles, J. Mater. Sci. 2006, 8035
(107) Jaroszczyk, T., Aerosol filtration in nanofiber filter media, American Filtration & Separations Society
(108) Adamiak, K., Aerosol deposition on an arbitrarily oriented single rectangularfiber in a uniform electric field, IEEE Trans. Ind. Appl. 1999, 35, 203-207
(109) Breschi, R.; Cuccuru, A.; Lanza, S., Scanning electron microscopy of ULPA and HEPA filtering papers
(110) Schreuder-Gibson, H.L., Gibson, P., Applications of electrospun nanofibers in current and future materials, U.S. Army Research, Development & Engineering Command Natick Soldier Center, Kansas Street, Natick, MA
(111) Bell, K., Polymer of 2-propenoic acid, butyl ester, diethenylbenzene, and
ethenylbenzene, National Industrial Chemicals Notification and Assessment Scheme, Full Public Report, 1991
(112) Calle, S.; Contal, P.; Thomas, D.; Bemer, D.; Leclerc, D., Description of the clogging and cleaning cycles of filter media, Powder Technol 2002, 123, 40-52